Systems, methods, and applications for radiotherapy treatment pre-planning

Abstract

Systems, methods, and application are provided for radiotherapy treatment pre-planning. Virtual contoured anatomical structures defining a target volume, non-target volumes of tissue, and a patient surface are used to derive tangential geometric reference structures that guide placement of candidate radiation delivery geometries. A simplified, geometry-based dose calculation model is applied to evaluate radiation delivery geometries to the irradiation of the non-target volumes. Iterative selection, constraint enforcement, and optional bolus and tuning structures support the generation of a treatment pre-plan that can be provided to an existing treatment planning system for dose optimization and delivery by a radiation therapy system.

Description

SYSTEMS, METHODS, AND APPLICATIONS FOR RADIOTHERAPY TREATMENT PRE-PLANNINGCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims priority to U.S. Provisional Patent Application No.63 / 774,979, titled “SYSTEMS, METHODS, AND APPLICATIONS FOR AUTOMATED RADIOTHERAPY TREATMENT PLANNING” and filed on March 20, 2025, the entire contents of which is incorporated herein by reference.BACKGROUND OF THE DISCLOSURE

[0002] The present disclosure relates to radiation therapy, and more particularly to computerized systems, methods, and software applications for planning radiation treatment procedures.

[0003] Radiotherapy is commonly used in the treatment of skin cancer and other superficial or near-surface malignancies. Depending on the clinical indication, radiation treatment may be delivered using photon (x-ray) beams or electron beams. Radiotherapy may be applied to a wide range of cancers, including basal cell carcinoma, squamous cell carcinoma, melanoma, Merkel cell carcinoma, cutaneous T-cell lymphoma, mycosis fungoides, various sarcomas, and other malignant or pre-malignant conditions.

[0004] Prior to delivering radiation therapy, a detailed computerized treatment plan can be created in preparation for treating the patient, where the detailed computerized treatment plan ensures that a prescribed dose of radiation is delivered to a target volume while minimizing dose to surrounding healthy tissue. For electron beam treatment planning, radiation is typically delivered using a single beam incident on the target volume. As a result, electron treatment planning is generally simpler and faster to perform, but is limited in its ability to conform the dose distribution to complex patient anatomy or irregular target geometries.

[0005] Photon beam treatment planning is generally more complex and involves multiple planning steps. These steps may include specifying the treatment prescription, defining the geometry of one or more radiation beams or beam arcs incident on the target, optimization to determine the intensity patterns of beams or arcs, optimizing beam intensities or fluence patterns, performing forward dose calculation to generate a three-dimensional (3D) spatial distribution of dose, and normalizing the dose to ensureadequate target coverage while respecting dose constraints for organs at risk. Because many of these steps are subjective in terms of input parameters and clinical judgment, photon treatment planning is typically performed as an iterative process. Completion of a clinically acceptable plan often requires several hours of work by a trained treatment planner, such as a dosimetrist or medical physicist.

[0006] While photon treatment planning is applicable to almost all radiotherapy treatment delivery platforms, including traditional C-arm linear accelerators and O-ring linear accelerators, certain delivery platforms, such as O-ring linear accelerators, do not support electron beam delivery and thus has motivated improved and more efficient methods for treatment planning that generate comparable or improved plan quality compared to electron treatment plans. In addition, the time and expertise required to generate high-quality treatment plans represent a significant resource burden in clinical practice. With increasing patient volumes and growing shortages of skilled healthcare professionals, there is a strong need for systems and methods that streamline and where possible, completely automate one or more aspects of the radiation treatment planning process. Such improvements may reduce planning time, improve plan consistency, and enable broader access to high-quality radiotherapy treatment.

[0007] The background herein is included solely to explain the context of the disclosure. This is not to be taken as an admission that any of the material referred to was published, known, or part of the common general knowledge as of the priority date.SUMMARY OF THE DISCLOSURE

[0008] According to an aspect, there is provided a system for generating a radiotherapy treatment pre-plan for the delivery of a dose of radiation to a target volume of a patient, the system comprising: processing hardware comprising at least one processor and associated memory, the memory storing instructions that, when executed by the at least one processor, cause the system to perform operations that include: obtaining a set of virtual, contoured anatomical structures, at least some of the contoured anatomical structures defining the target volume of the patient, a non-target volume of the patient, and a patient surface, receiving an input from the user that defines a radiotherapy treatment modality and at least one treatment plan parameter for the radiotherapy treatment pre-plan, defining, based on the target volume, a tangential geometric reference structure which approximates a tangent to at least part of a surface-facing portion of a contoured anatomical structure that defines the target volume, the surface-facing portion being oriented toward the patient surface, employing the selected radiotherapy treatment modality, the at least one treatment plan parameter, and the at least one tangential geometric reference structure to determine an initial set of candidate radiation delivery geometries that are defined relative to the tangential geometric reference structure, employing a simplified dose calculation model to estimate, for each candidate radiation delivery geometry of the initial set of candidate radiation delivery geometries, a respective dose estimate value that corresponds to an irradiation of the non-target volume of the patient, wherein the respective dose estimate value is derived based on an extent of a geometric interaction or overlap between i) a virtual projection of the candidate radiation delivery geometry through the contoured anatomical structures, and ii) the non-target volume within the contoured anatomical structures, and selecting a subset of candidate radiation delivery geometries from the initial set of candidate radiation delivery geometries for inclusion in an updated radiotherapy treatment pre-plan based on a reduction of dose delivered to the non-target volume, as represented by the respective dose estimate of each candidate radiation delivery geometry.

[0009] A further understanding of the functional and advantageous aspects of the disclosure can be realized by reference to the following detailed description and drawings.BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Embodiments are described with reference to the accompanying drawings. In the drawings, like reference numbers can indicate identical or functionally similar elements.

[0011] FIG. 1A shows a schematic diagram of an embodiment of the system for the generation of a radiotherapy treatment pre-plan.

[0012] FIG. 1B shows a flowchart of a method for generating the radiotherapy treatment pre-plan based on the system of FIG. 1A.

[0013] FIG. 2A shows a schematic diagram of a virtual representation of a target volume, patient surface, and tangential geometric reference structure.

[0014] FIG. 2B shows a flowchart of a method for determining tangent angles of a tangential geometric reference structure.

[0015] FIG. 3A shows a diagram of a virtual representation of a bolus structure, a target volume, patient surface, tangential geometric reference structure, and body contours.

[0016] FIG. 3B shows a flowchart of a method for iteratively estimating a dose estimate value for an initial set of radiation delivery geometries.

[0017] FIG. 4 shows a diagram of a set of radiation delivery beams positioned relative to virtual representations of a target volume, patient surface, and tangential geometric reference structure.

[0018] FIG. 5 shows a flowchart of a method for defining the positions of the beams in FIG. 4, relative to the tangential geometric reference structure.

[0019] FIG. 6 shows a diagram of a process for optimization the selection and placement of a subset of radiation delivery beams positioned relative to virtual representations of a target volume, patient surface, and tangential geometric reference structure.

[0020] FIG. 7 shows a flowchart of a method for iteratively estimating a dose estimate value of the beams in FIG. 6.

[0021] FIG. 8 shows a diagram of a radiation delivery arc positioned relative to virtual representations of a target volume, patient surface, and tangential geometric reference structure.

[0022] FIG. 9 shows a flowchart of a method for defining start and stop positions for the arc in FIG. 8, relative to the tangential geometric reference structure.

[0023] FIG. 10 shows a diagram of a process for optimization the selection and placement of a subset of radiation delivery arcs positioned relative to virtual representations of a target volume, patient surface, and tangential geometric reference structure.

[0024] FIG. 11 shows a flowchart of a method for iteratively estimating a dose estimate value of the arcs in FIG. 10.

[0025] FIG. 12 shows a schematic diagram of the projection of a candidate radiation delivery through a virtual representation of the target volume and non-target volume as part of an embodiment of the simplified dose calculation method.

[0026] FIG. 13 shows a flowchart of a method for iteratively estimating a dose estimate value based on the schematic in FIG. 12.

[0027] FIG. 14 shows a schematic diagram of an embodiment of the processing hardware of the system in FIG. 1A.DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] Various embodiments and aspects of the disclosure will be described with reference to details discussed below. The following description and drawings are illustrative of the disclosure and are not to be construed as limiting the disclosure. Numerous specific details are described to provide a thorough understanding of various embodiments of the present disclosure. However, in certain instances, well-known or conventional details are not described in order to provide a concise discussion of embodiments of the present disclosure.

[0029] As used herein, the terms “comprises” and “comprising” are to be construed as being inclusive and open ended, and not exclusive. Specifically, when used in the specification and claims, the terms “comprises” and “comprising” and variations thereof mean the specified features, steps or components are included. These terms are not to be interpreted to exclude the presence of other features, steps or components.

[0030] As used herein, the terms “about” and “approximately” are meant to cover variations that may exist in the upper and lower limits of the ranges of values, such as variations in properties, parameters, and dimensions. Unless otherwise specified, the terms “about” and “approximately” mean plus or minus 25 percent or less.

[0031] It is to be understood that unless otherwise specified, any specified range or group is as a shorthand way of referring to each and every member of a range or group individually, as well as each and every possible sub-range or sub-group encompassed therein and similarly with respect to any sub-ranges or sub-groups therein. Unless otherwise specified, the present disclosure relates to and explicitly incorporates each and every specific member and combination of sub-ranges or sub-groups.

[0032] As used herein, the term "on the order of", when used in conjunction with a quantity or parameter, refers to a range spanning approximately one tenth to ten times the stated quantity or parameter.

[0033] For the purpose of contextualizing the structure and operation of the systems, devices, and methods disclosed herein, headings are provided. The headings provided herein are for convenience only and do not interpret the scope or meaning of the claimedpresent technology. Embodiments under any one heading may be used in conjunction with embodiments under any other heading.

[0034] For simplicity and clarity of illustration, where considered appropriate, reference numerals may be repeated among the Figures to indicate corresponding or analogous elements. In addition, numerous specific details are set forth in order to provide a thorough understanding of the embodiment or embodiments described herein. However, it will be understood by those of ordinary skill in the art that the embodiments described herein may be practiced without these specific details. In other instances, well-known methods, procedures and components have not been described in detail so as not to obscure the embodiments described herein. It should be understood at the outset that, although exemplary embodiments are illustrated in the figures and described below, the principles of the present disclosure may be implemented using any number of techniques, whether they are currently known or not. The present disclosure should in no way be limited to the exemplary implementations and techniques illustrated in the drawings and described below.

[0035] Various terms used throughout the present description may be read and understood as follows, unless the context indicates otherwise: “or” as used throughout is inclusive, as though written “and / or”; singular articles and pronouns as used throughout include their plural forms, and vice versa; similarly, gendered pronouns include their counterpart pronouns so that pronouns should not be understood as limiting anything described herein to use, implementation, performance, etc. by a single gender; “exemplary” should be understood as “illustrative” or “exemplifying” and not necessarily as “preferred” over other embodiments. Further definitions for terms may be set out herein; these may apply to prior and subsequent instances of those terms, as will be understood from a reading of the present description. It will also be noted that the use of the term “a” or “an” will be understood to denote “at least one” in all instances unless explicitly stated otherwise or unless it would be understood to be obvious that it must mean “one”.

[0036] Modifications, additions, or omissions may be made to the systems, apparatuses, and methods described herein without departing from the scope of the disclosure. For example, the components of the systems and apparatuses may be integrated or separated. Moreover, the operations of the systems and apparatuses disclosed herein may be performed by more, fewer, or other components and themethods described may include more, fewer, or other steps. Additionally, steps may be performed in any suitable order. As used in this document, “each” refers to each member of a set or each member of a subset of a set.

[0037] The embodiments described herein are exemplary (e.g., in terms of materials, shapes, dimensions, and constructional details) and do not limit the claims appended hereto and any amendments made thereto. Persons skilled in the art will appreciate that there are yet more alternative implementations and modifications possible, and that the following examples are only illustrations of one or more implementations.

[0038] Unless defined otherwise, all technical and scientific terms used herein are intended to have the same meaning as commonly understood to one of ordinary skill in the art.

[0039] Despite advances in radiotherapy delivery technology, the process of treatment planning remains time-consuming and highly dependent on the skill and clinical judgement of the individual making the plan, making the process of treatment planning particularly difficult to standardize. The treatment of skin and superficial lesions presents particular set of technical challenges. Unlike radiation therapy planning for deep-seated tumors, the planning of radiation therapy treatments for skin and superficial lesions includes a target volume (for irradiation) that is positioned at or near a patient surface.

[0040] Photon-based treatment planning for such indications often requires manual selection of different input parameters (e.g., beam geometries, field sizes, margins, bolus configurations, etc.). For skin and superficial lesions, the planning often requires a bolus of radiation delivered to the target volume needs to be controlled so that dose deposition occurs only at shallow depths. Careful selection of beam angles may also be required to accommodate irregular, patient-specific body contours near the target volume. The selection of each of these different input parameters can significantly impact the resulting dose distribution and in practice, the selections are primarily informed by the planner’s own experience and trial-and-error iteration.

[0041] Existing clinical workflows for the planning of radiation therapy treatments do include treatment planning software systems, such as the Varian Eclipse, which can automate some aspects of the planning process. However, the operation of these systems requires many labor-intensive, time-consuming manual processes (e.g., the contouring of uniform -thickness bolus structures that conform to the patient’s anatomy and treatment region, adjustment of patient body contours to integrate the bolus whilemaintaining accurate original structure for dosimetric reporting, the creation of supplementary optimization structures (e.g., tuning rings or expanded target volumes) for dose modulation, the visual identification and iterative refinement of optimal beam angles based on the target volume and surface geometry). These highly manual processes within existing treatment planning systems can require between 30 and 90 minutes per plan and is heavily reliant on planner expertise. This can result in significant variability in both workflow efficiency and clinical outcomes across practitioners and institutions. The lack of automation and standardization in this process underscores a need for improved methods and systems.

[0042] The challenges in radiation treatment planning are compounded when photon-based techniques are used to approximate or replace electron treatments, such as on treatment platforms that do not support electron beam delivery. Existing planning workflows provide limited guidance or automation for these scenarios, often requiring repeated manual adjustments and dose recalculations. As a result, planning efficiency is reduced and consistency between different planning procedures is difficult to achieve.

[0043] The computerized system, methods, and application of the present disclosure seek to address these and other challenges by providing a radiotherapy treatment planning procedure that is configured to automate aspects of the treatment planning in order to reduce planner burden and generate clinically acceptable treatment plans with reduced manual intervention.

[0044] Embodiments of the present disclosure provide for a system comprising processing hardware, as well as computerized methods and applications for the automated or semi-automated preparation of a radiotherapy treatment pre-plan. The radiotherapy treatment pre-plan can be utilized by an existing treatment planning system to expedite the generation of a final radiotherapy treatment plan intended for the delivery of a dose of radiation to a target volume of tissue of the patient. The computerized system, methods, and applications provided herein can streamline the pre-planning process performed by radiologists or radiology technicians. Additionally, the automated or semi-automated generation of the radiotherapy treatment pre-plan can provide for more robust and consistent generation of the radiotherapy treatment plans.

[0045] The present inventor, having encountered the complexity and challenges in using modern radiation treatment planning systems to plan radiation treatment procedures, realized that workflow efficiency could be improved by providing a user-interface-driven preliminary planning subsystem that i) accepts input and / or guides the user-definition of treatment planning baseline data and parameters, such as, but not limited to, dose, fraction, volumetric image data, structure sets, boluses, radiation therapy treatment beam energy, radiation therapy treatment modality, ii) employs the baseline data and parameters to generate a beam arrangement plan that mitigates the irradiation of the non-target volume of tissue (or another suitable measure of peripheral tissue exposure), and iii) interfaces with a treatment planning system and provides, to the treatment planning system, the beam arrangement plan and other parameters suitable for subsequent detailed planning steps by the treatment planning system, including dose optimization and normalization. It has been found that such an approach to pre-planning can streamline the workflow and generate useful and accurate radiation therapy plans with a substantial reduction in time and resources, both due to factors including i) ease of use by the user and (ii) a reduction in the computing resources required by the treatment planning system due to the provision of the preliminary beam arrangement plan and other inputs that can be readily employed by the treatment planning system to generate a final plan.

[0046] Referring to FIG. 1A, an example embodiment of the system 100 of the present disclosure is schematically illustrated. The system 100 includes processing hardware 500. In the example embodiment of FIG. 1 A, one or more process applications (automation applications) 130 are provided for by the system 100. In certain embodiments, the one or more process applications detailed herein may be stored on the processing hardware 500 of the system and executed by such hardware to perform the methods as part of the preparation of the radiotherapy treatment pre-plan. The process applications 130 can be stored on a memory of the processing hardware 500, and executable by processor(s) of the processing hardware 500. Furthermore, the computer-implemented methods disclosed herein may be programmed as features of the applications 130 which are executable on the processing hardware 500. Further details of the processing hardware 500 of the system 100 are provided below with reference to FIG. 14.

[0047] FIG. 1A also illustrates an example workflow for the generation of the radiotherapy treatment pre-plan. As shown in the figure, the workflow may comprise steps such as, but not limited to: the addition of a new treatment course, plan and prescription, the generation of automated planning structures that will assist in effectivedose optimization, the automatic generation of a bolus based on the target volume, the automatic and strategic placement of beams or beam arcs, dose optimization to determine x-ray intensity patterns (for intensity modulated radiotherapy, or IMRT) or beam apertures (for Volumetric Modulated Arc Therapy, VMAT), forward dose calculation, and dose normalization to generate the automated radiotherapy plan.

[0048] In at least some example embodiments disclosed herein, the process application is an automatic process application that automatically performs, based on user input, at least some of the process steps described herein for generating the radiation treatment pre-plan.

[0049] As shown in FIG. 1A, the one or more process applications 130 are generally configured to prepare the radiotherapy treatment pre-plan by receiving inputs 110 from a user (e.g., an operator) that can define various treatment plan parameters, such as, but not limited to, dose, fraction, volumetric image data, structure sets, bolus parameters, radiation therapy treatment beam energy, radiation therapy treatment modality. The one or more process applications 130 are also configured to employ the treatment plan parameters to implement the methods described herein for generating the radiotherapy treatment pre-plan. The at least one treatment plan parameter may further define aspects of the overall radiation treatment plan such as the treatment course, plan, and prescription. The treatment course, plan, and prescription can include information such as a total treatment dose, a number of fractions that the total treatment dose is to be split up into, and information about one or more target volumes that are to be targeted by radiation treatment. Once the treatment course, plan, and prescription have been input by the user, the process application may send commands to the treatment planning system to add the treatment course, plan, and prescription.

[0050] In embodiments, the process application is configured to prompt a user to add at least one pre-procedural image dataset of pre-procedural image data (such as CT scan data) to the application. The pre-procedural image data can include medical imaging acquired prior to treatment planning. The pre-procedural image data may include computed tomography (CT) data, magnetic resonance imaging (MRI) data, positron emission tomography (PET) data, or combinations thereof. The process application can display the pre-procedural image data via the display device(s) or can send the pre-procedural image data to the treatment planning system for display.

[0051] As shown in FIG. 1A, the process application may also use the body contours and the target volume to define one or more planning structure geometries. The process application can then add the planning structure geometries to the treatment planning system. The planning structure geometries generated by the process application can include, for example, a tangent to the bolus. Further details on the use of a tangent as a planning structure geometry is described below with reference to FIG. 2.

[0052] Referring again to FIG. 1A, the process application may prompt the user to select a treatment type (e.g., IMRT or VMAT) and to define various treatment details, where these treatment details may vary depending on the treatment type. For example, for an IMRT treatment, the user may be prompted to define a number of beams, a beam separation angle, and a method of beam placement. Alternatively, for an VMAT treatment, the user may be prompted to define a number of arcs, and a method of arc placement. From the input treatment type and other treatment details (e.g., beam or arc placement), a preliminary beam configuration for one or more treatment beams is defined.

[0053] In an additional embodiment also shown in FIG. 1A, the system 100 and process applications 130 are configured to interface with an existing treatment planning system 140 and can provide, to the treatment planning system 140, the radiotherapy treatment pre-plan (including the set of radiation delivery geometries) and other suitable parameters of the treatment plan parameters for subsequent detailed planning steps carried out via the treatment planning system 140 (e.g., dose optimization and dose normalization).

[0054] In the example embodiment in FIG. 1A, the process application 130 interfaces with the existing treatment planning system 140, where the existing treatment planning system 140 may be implemented via treatment planning software, and where the process application 130 interfaces with the treatment planning system 140 via a suitable communication protocol (e.g., via an application-program interface (API) of the treatment planning system 140). The process application 130 can provide output data / information to the treatment planning system 140 and receive treatment planning system-generated data / information from the treatment planning system 140. The process application may also prompt the treatment planning system 140 to perform a dose normalization process, and / or the dose normalization process may proceed following the initial dose optimization step performed by the treatment planning system. The process application may promptthe user to input dose normalization parameters, and then send the dose normalization parameters on to the treatment planning system and calls a forward dose calculation and dose normalization calculation from the treatment planning system. The process applications as described herein can be applied for planning radiation treatment for several types of radiation-treatable diseases.

[0055] In embodiments, the treatment planning system can include various known applications for radiation planning, such as the Varian Eclipse treatment planning system using the Eclipse API (ESAPI), described in the document “Eclipse Photon and Electron Algorithms Reference Guide”, published by Varian Systems with document ID P1008611-003-C in 2015, which is incorporated herein by reference in its entirety.Introduction to Virtual Constructs

[0056] In at least some embodiments, the generation of the radiotherapy treatment pre-plan uses a set of virtual contoured anatomical structures that are associated with a patient. The virtual contoured anatomical structures are provided as virtual representations of patient anatomy and are defined within a coordinate space used by the radiotherapy treatment pre-planning and treatment planning workflow.

[0057] The virtual contoured anatomical structures may be derived either directly or indirectly from the pre-procedural image data for the patient. In some embodiments, the virtual contoured anatomical structures are pre-defined or pre-segmented within the pre-procedural image data and may be selected by the user within the process application. In other embodiments, the user may define, modify, or refine one or more of the virtual contoured anatomical structures using the process application.

[0058] In certain embodiments, the contouring of at least some of the virtual contoured anatomical structures defines virtual representations of the target volume of the patient intended to receive the dose of radiation, one or more non-target volumes of the patient (representing non-target tissue for which radiation exposure is to be limited), and at least one patient surface. The virtual contoured anatomical structures can act as reference bases for the subsequent pre-planning operations described within the system, methods and process applications of the present disclosure, including the definition of geometric reference structures, generation of candidate radiation delivery geometries, and evaluation of candidate radiation delivery geometries using a simplified dose calculation model.

[0059] In additional embodiments, the virtual representations of the one or more nontarget volumes also include virtual representations of one or more organs at risk (OARs), normal tissue regions, or other anatomical regions for which radiation exposure is to be minimized during generation of the radiotherapy treatment pre-plan.

[0060] As provided above, the radiotherapy treatment pre-plan also defines a set of radiation delivery geometries. These radiation delivery geometries can be defined based on the contoured anatomical structures and at least one geometric reference structure that is derived from the contoured anatomical structures. Details of the derivation of the at least one geometric reference structure are provided in further detail below.

[0061] Generally, the set of radiation delivery geometries are formed as virtual, structured representations of the means by which a dose of radiation will be delivered relative to the patient anatomy (e.g., the target volume, the one or more non-target volumes and the patient surface). The set of radiation delivery geometries can be determined independent of a calculated radiation dose distribution. The set of radiation delivery geometries are, in some embodiments, specifically selected to reduce an amount of irradiation of the non-target volume of tissue (e.g., tissue that is peripheral to the target tissue). Other suitable metrics of non-target volume irradiation exposure can also be employed when selecting the set of radiation delivery geometries.

[0062] During the generation of the radiotherapy treatment pre-plan, an initial set of candidate radiation delivery geometries are defined and assessed so as to reduce this non-target volume of irradiated (non-target) tissue. Each candidate radiation geometry of this initial set is defined relative to the geometric reference structure or at one or more offsets relative to the geometric reference structure (e.g., at one or more angular offsets relative to a tangential geometric reference structure).

[0063] In an additional embodiment, a simplified dose calculation model is applied to assess the plurality of candidate radiation delivery geometries in generating the radiotherapy treatment pre-plan. The simplified dose calculation model is configured to estimate / approximate a dose of radiation delivered to the non-target tissue by each candidate radiation delivery geometry of the plurality of candidate radiation delivery geometries, where this estimate / approximation is based at least one geometric characteristic of the candidate radiation delivery geometries and a relationship of the radiation delivery geometry to other anatomical structures of the patient defined within the contoured anatomical structures. The application of the simplified dose calculationmodel to the generation of the set of radiation delivery geometries in the treatment preplan can reduce the computing time and power required to both i) generate the radiotherapy treatment pre-plan and ii) generate the final radiotherapy treatment plan via the treatment planning system.

[0064] Referring to FIG. 1 B, there is provided a first embodiment of a computer-implemented method for generating the radiation treatment pre-plan. In this embodiment, the method comprises a step S110 of obtaining a set of the virtual, contoured anatomical structures, where at least some of the contoured anatomical structures define the target volume of the patient, the non-target volume of the patient, and the patient surface. The method also comprises a step S120 of receiving an input from the user that defines the radiotherapy treatment modality to be used, as well as treatment plan parameters for the radiotherapy treatment plan and radiotherapy treatment modality. The method comprises a step S130 of defining, based on the target volume, a tangential geometric reference structure that approximates a tangent to at least part of a surface-facing portion of a contoured anatomical structure that defines the target volume, with the surface-facing portion being oriented toward the patient surface. In addition, the method comprises a step S140 of employing the selected radiotherapy treatment modality, the at least one treatment plan parameter, and the at least one tangential geometric reference structure to determine an initial set of candidate radiation delivery geometries (defined relative to the tangential geometric reference structure), and a step S150 of employing a simplified dose calculation model to estimate, for each candidate radiation delivery geometry of the initial set, a respective dose estimate value. The respective dose estimate value is an approximation of the irradiation of the non-target volume of the patient by the given candidate radiation delivery geometry. Lastly, the method comprises a step S160 of selecting a subset of candidate radiation delivery geometries from the initial set of candidate radiation delivery geometries for inclusion in an updated radiotherapy treatment pre-plan. In the step S160, the subset of candidate radiation delivery geometries can be selected based on a reduction of the dose that would be delivered to the non-target volume, as represented by the respective dose estimate of a given candidate radiation delivery geometry.

[0065] In an additional embodiment, the simplified dose calculation model is configured to generate the optimal subset of candidate radiation delivery geometries (based on an initial set of candidate radiation delivery geometries) in order to pass thesubset as part of radiotherapy treatment pre-plan to the treatment planning system. The treatment planning system can then generate a final radiotherapy treatment plan based, at least in part, on the radiotherapy treatment pre-plan.Virtual Contoured Anatomical Structures

[0066] As provided above, embodiments of the virtual contoured anatomical structures can define the target volume of the patient, the one or more non target volumes, and the representation of the patient surface. The virtual contoured anatomical structures may be derived from pre-procedural image data, such as volumetric medical imaging acquired prior to treatment planning. The pre-procedural image data may include computed tomography (CT) data, magnetic resonance imaging (MRI) data, positron emission tomography (PET) data, or combinations thereof.

[0067] In some embodiments, the virtual contoured anatomical structures may be represented solely as geometric or coordinate-based data without requiring the display of the underlying image data during the generation of the radiation treatment pre-plan. In at least some other embodiments, the virtual contoured anatomical structures may be represented by underlying pre-procedural image data of the patient. At least a portion of the pre-procedural image data may be presented on a display of the system to facilitate one or more steps in the generation of the radiation treatment pre-plan (e.g., an input by the user associated with the display image data).

[0068] In an additional embodiment, the virtual contoured anatomical structures are provided as part of one or more structure sets, such as structure sets formatted in accordance with a DICOM standard (or another suitable data format). The structure sets may include pre-segmented or pre-defined volumetric representations of anatomical regions, including the target volume, body contours of the patient, and the one or more non-target volumes. The structure sets may be imported from an external system, generated automatically, or previously prepared as part of a clinical workflow. The user can select a body structure set, and one or more target volumes within the structure set. Using the generated body contours, the structure set, and the target volume selected by the user, the process application can define the target volume relative to the body contours within the pre-procedural image data. Once the target volume is defined in the pre-procedural image data, a region for the bolus around the one or more target volumes can be defined.

[0069] The virtual contoured anatomical structures may be defined or refined using user input, automatic processes, ora combination thereof. In some embodiments, a user may manually define or adjust at least one of the contoured anatomical structures. In other embodiments, the user may select at least one pre-defined structure from a structure set without modifying the underlying contours. In still other embodiments, at least one of the virtual contoured anatomical structures may be generated automatically using segmentation algorithms or predefined rules, without requiring direct user interaction.Virtual Geometric Reference Structures

[0070] As provided above, the at least one geometric reference structures is derived from the contoured anatomical structures and used, at least in part, as a basis for defining the set of radiation delivery geometries during the generation of the radiotherapy treatment pre-plan. The at least one geometric reference structure is generally derived relative to, or using, the contours of the target volume and the patient surface in the contoured anatomical structures.

[0071] In an embodiment, the at least one geometric reference structure includes a tangential geometric reference structure that defines a tangent relative to some aspect(s) of the contoured anatomical structures of the target volume and / or the patient surface. The tangential geometric reference structure can be implemented using various types of virtual constructs. For example, the tangential geometric reference structure may comprise a tangent line, a tangent plane, or another suitable tangent-based virtual construct.

[0072] Generally, the tangential geometric reference structure is used for the virtual placement of both the initial set of radiation delivery geometries and the updated subset of radiation delivery geometries during the generation of the radiotherapy treatment pre-plan. In particular, the tangential geometric reference structure may be used as a geometric basis for positioning or orienting candidate radiation delivery geometries (e.g., orienting the candidate radiation delivery geometries angularly relative to the tangent plane that defines the tangential planning structure).

[0073] Details of the tangential geometric reference structure and example embodiments and derivations of the tangential geometric reference structure are described below with reference to FIGS. 2A to 2B.

[0074] In some embodiments, the tangential geometric reference structure is generated automatically, with the system being configured to automatically define the tangential geometric reference structure without requiring manual definition of tangential orientations by a user. In some other embodiments, the tangential geometric reference structure is generated semi-automatically or completely manually and involves a user manually defining at least some aspects of the tangential geometric reference structure.

[0075] In an embodiment where the at least one geometric reference structure includes the tangential geometric reference structure, the tangential geometric reference structure is configured to represent, characterize or approximate an angular orientation of a surface of a target volume relative to the patient. In this embodiment, the tangential geometric reference structure provides a geometric representation of an approximate tangent associated with a surface-facing portion of the target volume (e.g. a surfacefacing portion of the contour defining or enclosing the target volume) and serves as a spatial and directional reference for defining angular orientations of radiation delivery geometries relative to the target volume and the patient surface.

[0076] In some example embodiments, the tangential geometric reference structure is defined such that a radiation beam directed along the tangential geometric reference structure (or angled slightly relative to the tangential reference structure, e.g. at an angle of less than 2, 5 or 10 degrees) and passing through the target volume would not irradiate one or more deeper anatomical structures that are further from the patient surface than the target volume.

[0077] In some embodiments, a surface facing portion of the target volume (e.g. the surface-facing portion of the contour defining or enclosing the target volume) is proximate (e.g. nearby, adjacent to, closest to) the patient surface, and the tangential geometric reference structure is defined to pass through or near a location on the patient surface and has an orientation determined based on a local surface geometry of a part of the surface facing portion of the target volume (e.g. a surface-facing portion of the contour defining or enclosing the target volume) that is closest to the patient surface.

[0078] In another alternate embodiment, the tangential geometric reference structure is defined to approximate a tangent to at least a portion of a surface-facing region of a contoured anatomical structure that defines the target volume (e.g. a surface-facing portion of the contour defining or enclosing the target volume). In this embodiment, the surface-facing region of the target volume is generally oriented toward the patientsurface. An example embodiment of the use of a tangent plane as the tangential reference structure is illustrated in the example embodiment of FIG. 2A.

[0079] In the specific embodiment shown in FIG.2A, the tangential geometric reference structure 200 comprises the tangent plane or line 200a. This example further illustrates virtual representations of the contoured anatomical structures, including a virtual representation of the target volume 210, defined by a virtual isocenter 212 and positioned proximate to a virtual representation of patient surface 220. A surface-facing segment 216 of the virtual representation of the target volume 210 is depicted, with tangent plane 200a aligning tangentially to a portion of the virtual representation of the surface-facing area 220.

[0080] In additional embodiments, the tangential geometric reference structure (e.g., a tangent plane or line) provides a representative geometric structure that is offset from the surface contour of the target volume (e.g., positioned closer to a treatment isocenter of the target volume or another reference location) or offset relative to the patient surface. The offset of the tangential geometric reference structure may be selected to account for treatment setup considerations or machine geometry and may result in a tangential geometric reference structure that represents an offset approximation of a tangent rather than an exact mathematical tangent to the target volume boundary.

[0081] The use of a tangential geometric reference structure in the positioning or orienting of the candidate radiation delivery geometries can provide for several benefits in the delivery of the dose of radiation. Using radiation delivery geometries that are tangentially oriented can result in a reduced volume of normal tissue being traversed by radiation, concentrating the dose of radiation at a superficial depth from the patient surface that is near the target volume. The tangentially oriented radiation delivery geometries can also limit the irradiation of deeper anatomical structures that are further from the patient surface. In some embodiments, the tangential oriented radiation delivery geometries produce glancing angles of radiation incidence relative to the patient surface, which can further contribute to sparing of non-target tissue.

[0082] The use of the tangential geometric reference structure may be particularly beneficial when the system, methods, and process application are employed for planning radiation treatment for skin cancer or other superficial lesions. In the contoured anatomical structures for these types of targets, the target volume is often located near the patient surface, and tangentially oriented radiation delivery geometries can beespecially effective in achieving adequate target coverage of the target volume while limiting the dose to underlying tissue.

[0083] In at least some embodiments of the geometric reference structures, the virtual contoured anatomical structures include a three-dimensional representation of the target volume, and the tangential geometric reference structure is defined by reducing the three-dimensional representation of the target volume to a two-dimensional geometric analysis. This two-dimensional geometric analysis can be performed on at least one representative image slice selected from a set of pre-procedural image slices that encompass the target volume. The representative image slice may be selected based on anatomical relevance, geometric characteristics of the target volume, or predefined selection criteria.

[0084] For example, the tangential geometric reference structure can include a tangent line or a tangent plane that is tangent to the surface-facing portion of the target volume, and the contoured anatomical structures defining the target volume can be derived from the pre-procedural CT image data of the patient. The pre-procedural CT image data can include axial CT image slices, and the target volume can be represented as a three-dimensional structure defined by a set of closed contours across the axial CT slices, with patient orientation information obtained from image metadata in the CT image data.

[0085] In one such embodiment, the tangent line or tangent plane is determined as either a mean or aggregate tangent computed across a plurality of CT image slices from the pre-procedural CT image data. The plurality of CT image slices span a superiorinferior extent of the target volume. The aggregate tangent is either derived from all image slices containing the target volume or from a selected subset of slices chosen to represent the geometric characteristics of the target volume along the superior-inferior axis.

[0086] In an alternate example embodiment, the tangential geometric reference structure includes a tangent line or tangent plane that is determined based on a single representative image slice corresponding to a superior-inferior midpoint of the target volume. The representative image slice is defined through the midpoint of the target volume along the superior-inferior direction, and the tangent line or tangent plane is derived from the contour geometry of the target volume on that slice. This approachprovides a simple yet robust approximation of tangential orientation for target volumes that are elongated or distributed across multiple image slices.

[0087] In at least some other embodiments of the geometric reference structures, the tangential geometric reference structure is defined by at least one tangent angle associated with the target volume. The at least one tangent angle can include a pair of opposing tangent angles defined as gantry angles (a^ a2), where the pair of opposing tangent angles correspond to two tangential radiation delivery orientations relative to the target volume.

[0088] In one such embodiment where the at least one tangent angle defines the tangential geometric reference structure, the at least one tangent angle is determined using a tangent angle algorithm. The tangent angle algorithm can, in certain embodiments, be configured to reduce the three-dimensional representation of the target volume to a two-dimensional geometric representation and to process the two-dimensional representation to determine the at least one tangential angle. The tangent angle algorithm may be implemented using various geometric techniques and is not limited to a single computational approach.Example Implementation of Tangent Angle Algorithm

[0089] In another example embodiment, the tangent angle algorithm reduces the three-dimensional representation of the target volume to a two-dimensional point set corresponding to a contour of the target volume on a representative image slice, such as an axial CT slice. The algorithm then extracts a two-dimensional point set representing the target volume contour on the selected slice, determines geometric extents of the point set, and computes the at least one tangent angle of the tangential geometric reference structure based on geometric extrema of the two-dimensional point set. In this embodiment, the algorithm adapts its computation based on whether the cross-section of the target volume is wider than it is tall (X-extent greater than Y-extent) or taller than it is wide (Y-extent greater than or equal to X-extent).

[0090] FIG. 2B provides a flowchart of an example implementation of this tangent angle algorithm. In this embodiment, the tangent angle algorithm includes a step S210 where a representative axial image slice is selected at a mid-Z position of the target volume, subject to a minimum point threshold to ensure sufficient contour data. The method also includes a step S220 where a two-dimensional point set corresponding to the target volume contour on the selected slice is extracted, and X- and Y-directiongeometric extents of the point set are computed. The method further includes a step S230 where, if the X-direction extent exceeds the Y-direction extent, indicating a wider target cross-section, the algorithm identifies contour points at minimum and maximum X positions and computes a tangent angle based on the difference in their Y-coordinates, and a step S240 where, if the Y-direction extent equals or exceeds the X-direction extent, indicating a taller target cross-section, the algorithm identifies contour points at minimum and maximum Y positions and computes a tangent angle based on the difference in their X-coordinates. Finally, the method includes a step S250, where the computed tangent angle is transformed into IEC61217 gantry coordinate space based on the recorded patient orientation. Example 1 provides further details on the mathematical calculations involved in this example embodiment of the algorithm.

[0091] This embodiment of the tangent angle algorithm automatically selects different geometric reference points based on an aspect ratio of a cross-section of the target volume. This adaptive behavior enables robust determination of tangential orientation for both relatively flat (wide) target geometries and elongated (tall) target geometries, thereby improving the reliability and consistency of tangential geometric reference structure generation across a wide range of anatomical presentations.

[0092] In an additional embodiment, the tangent angle algorithm is further configured such that, after the at least one tangent angle for the tangential geometric reference structure has been determined, an optional angular offset (A) is applied to the tangential geometric reference structure. The angular offset is used to adjust the position and tangential orientation of radiation delivery geometries to achieve a desired clinical effect, such as positioning radiation delivery geometries slightly above the patient surface or directing radiation delivery geometries slightly into the patient’s tissue. The direction of the angular offset, whether clockwise or counterclockwise relative to a base tangent angle, may be selected based on geometric analysis, physical measurement, or a combination thereof.

[0093] In an example embodiment, an offset direction of the angular offset can be determined for a given tangent angle ai by generating a pair of test radiation delivery geometries at angular positions (ai + A) and (ai - A). For each test radiation delivery geometry, a geometric or physical measurement associated with the patient anatomy is then evaluated.

[0094] In another example embodiment, a source-to-surface distance (SSD) corresponding to each test radiation delivery geometry is determined using a treatment planning system interface. The direction of the angular offset is then selected based on a sign or value of the offset parameter. For example, when a positive offset is specified, corresponding to radiation delivery geometries positioned away from the patient surface, the direction associated with a larger SSD may be selected. Conversely, when a negative offset is specified, corresponding to radiation delivery geometries positioned further into tissue, the direction associated with a smaller SSD may be selected.Alternate Implementations of Tangent Angle Algorithm

[0095] Various alternative methods can be applied as part of the tangent angle algorithm to determine tangent angles and thereby define a tangential geometric reference structure, such as a tangent line or tangent plane. The following paragraphs provide various example embodiments of methods for the tangent angle algorithm.

[0096] In one example embodiment, the at least one angle is determined using a principal component analysis (PCA)-based version of the tangent angle algorithm, applied to a three-dimensional mesh representation of the target volume. In this embodiment, vertices of a three-dimensional mesh corresponding to the target volume are treated as a point cloud. PCA is performed on the point cloud to determine a set of orthogonal principal axes corresponding to directions of maximum geometric variance of the target volume.

[0097] A first principal component, representing a direction of greatest geometric extent of the target volume, defines a major axis of the target. Tangent angles may then be derived by projecting the major axis onto a transverse plane and computing a direction perpendicular to the projected major axis, which corresponds to an approximate tangential approach relative to a surface-facing portion of the target volume. This PCA-based method utilizes the full three-dimensional geometry of the target volume rather than relying on a single image slice and can naturally accommodate targets that are tilted, oblique, or irregularly shaped. The PCA-based approach is robust to irregular contours because all mesh vertices are considered simultaneously and is computationally efficient.

[0098] In another example embodiment, the at least one tangent angle is determined using an embodiment of the tangent angle algorithm that derives a surface normal estimation from a triangulated mesh representation of the target volume. In thisembodiment, a mesh triangulation of the target volume is analyzed to compute surface normal vectors associated with individual mesh elements, such as triangular facets. Surface normal vectors corresponding to an outward -facing portion of the target volume, for example a portion oriented toward the patient surface, may be identified and aggregated. An average or dominant surface normal may then be computed from the selected surface normals to represent a prevailing surface orientation of the target volume. The at least one tangent angle is subsequently determined as a gantry angle that is perpendicular to the dominant surface normal when projected onto a transverse plane. This surface-normal-based method directly reflects the geometric intent of tangential radiation delivery, in which radiation is delivered substantially parallel to a surface of the target volume. By relying on local surface orientation rather than global shape statistics, this approach can handle non-convex target geometries and cases in which the patient surface curves away from the target volume.

[0099] In another example embodiment, the at least one tangent angle is determined using an embodiment of the tangent angle algorithm that is based on a convex-hull geometric analysis of two-dimensional contour points extracted from one or more representative image slices containing the target volume. A convex hull is computed from the contour points, and a rotating-calipers algorithm is applied to identify a minimum-width direction of the convex hull. The direction corresponding to the smallest projected width of the target volume is selected, and the tangent angles are defined as orientations perpendicular to that direction. This embodiment provides a mathematically optimal tangential orientation in two dimensions and accommodates irregular target shapes through convexification of the contour geometry.

[0100] In yet another example embodiment, the at least one tangent angle is determined using an embodiment of the tangent angle algorithm that is based on a center-of-mass approach. A centroid of the target volume and a centroid of a patient body volume are computed from the contoured anatomical structures. A vector extending from the body centroid to the target-volume centroid defines a depth direction of the target within the patient. Tangent angles are then defined as gantry angles perpendicular to the depth direction when projected onto a transverse plane. This embodiment provides a computationally simple and intuitive approximation of tangential orientation based on relative target position within the patient anatomy.

[0101] In another example embodiment, the at least one tangent angle is determined using an embodiment of the tangent angle algorithm that is based on an iterative, dose-driven optimization approach. Starting from an initial tangent angle estimate, candidate angles are evaluated by computing a dose distribution or dose estimate and adjusting the angle to reduce a cost function associated with non-target irradiation. The optimization process iteratively refines the tangent angle until a termination condition is satisfied. This embodiment directly optimizes a clinically relevant dose-based metric and may be extended to optimize multiple beam angles simultaneously, albeit with increased computational cost.

[0102] In yet another example embodiment, the at least one tangent angle is determined using an embodiment of the tangent angle algorithm that uses ray-tracing from virtual radiation source positions. For each candidate gantry angle, a ray is traced from a source position through a three-dimensional representation of the patient anatomy and tested for intersection with a target-volume mesh. Tangent angles are identified as angles at which the ray transitions between intersecting and non-intersecting the target volume, corresponding to geometric tangency. This embodiment models beam-target interaction directly, accounts for beam divergence, and can be extended to non-coplanar delivery geometries.

[0103] In another example embodiment, the at least one tangent angle is determined using an embodiment of the tangent angle algorithm that uses Fourier-descriptor analysis of a two-dimensional contour of the target-volume. The contour is parameterized as a closed curve and decomposed into Fourier series components. Low-order Fourier coefficients capture dominant shape characteristics, including elongation and orientation. A major axis orientation is derived from the Fourier coefficients, and tangent angles are defined as orientations perpendicular to the major axis. This embodiment provides a smooth, noise-resistant characterization of target shape and supports derivation of tangential orientation from dominant geometric features.

[0104] In still another example embodiment, the at least one tangent angle is determined using an embodiment of the tangent angle algorithm that uses a multi-slice consensus approach. Tangent angle estimates are computed independently on each image slice containing the target volume using one or more of the described tangent-estimation methods. The per-slice estimates are then combined to produce a consensus tangent angle, for example using an arithmetic mean, weighted mean,median, or voting scheme. This embodiment incorporates geometric information from the full three-dimensional extent of the target volume and provides robustness against slice-specific irregularities or artifacts.Radiation Delivery Geometries

[0105] As provided above, the generation of the radiotherapy treatment pre-plan includes the defining of the initial set of candidate radiation delivery geometries. The initial set of candidate radiation delivery geometries defined during radiotherapy treatment pre-planning may include various different types of radiation delivery configurations, depending on the selected radiotherapy treatment modality. For example, the initial set of candidate radiation delivery geometries may define a set of discrete radiation delivery beams, or a set of continuous radiation delivery arcs. In embodiments where the selected radiotherapy treatment modality comprises intensity-modulated radiation therapy (IMRT), the discrete radiation delivery beams may include a plurality of fixed-angle beams. In embodiments where the selected radiotherapy treatment modality comprises volumetric modulated arc therapy (VMAT), the set of continuous radiation delivery arcs may include one or more radiation delivery arcs defined by corresponding angular extents.

[0106] The set of initial candidate radiation delivery geometries may be generated at, or angularly relative to, the at least one geometric reference structure (e.g., the tangential geometric reference structure), and may be distributed across a range of angular orientations or extents based on the treatment planning parameters. The set of initial candidate radiation delivery geometries are not meant to represent a final treatment plan, but instead form a pool of geometries that may be assessed and screened for suitability.

[0107] Example methods for generating both the initial set of radiation delivery geometries and the updated set of radiation delivery geometries are provided below with reference to FIGS. 4 to 11. Specifically, example methods for generating the initial set of radiation delivery geometries are provided below with reference to FIGS. 4 and 5, and FIGS. 8 and 9.

[0108] Referring to FIG. 4, there is provided an example embodiment where the radiotherapy treatment modality selected by the input of the user comprises IMRT. In at least some embodiments where the radiotherapy treatment modality is IMRT, the initial set of radiation delivery geometries includes a set of fixed-angle radiation delivery beams 412 (shown in FIG. 4). In at least some embodiments where the selected radiotherapy treatment modality is IMRT, the user defined input (as part of the at least one treatmentplan parameter) sets various parameters of the IMRT protocol. These parameters can include, for example, a desired number fixed-angle radiation delivery beams, a minimum separation angle between adjacent fixed-angle radiation delivery beams of the set of fixed-angle radiation delivery beams, and an angular offset relative to the tangential geometric reference structure.

[0109] In an embodiment such as shown in FIG. 4, the initial set of radiation delivery geometries for an IMRT protocol are generated by placing the set of fixed angle radiation delivery beams 412 at angular positions defined relative to the tangential geometric reference structure 200. In the specific embodiment shown in FIG. 4, the fixed-angle radiation delivery beams 412 include a first pair of fixed-angle radiation delivery beams 412a, 412b that are placed at angular positions defined by the angular offset relative to the tangent plane 200a for the target volume 210. The first pair of fixed-angle radiation delivery beams 412a, 412b are positioned on opposing sides of the tangent plane 200a. The fixed-angle radiation delivery beams 412 also include a number of additional fixed-angle radiation delivery beams 412 that are disposed at angular positions that are spaced from one another and from the first pair of fixed-angle radiation delivery beams 412a, 412b by at least the minimum separation angle (9).

[0110] FIG. 5 provides a flowchart of an example method for determining a beam arrangement for the plurality of fixed-angle radiation delivery beams in the IMRT protocol illustrated in FIG. 4. In this method for determining the beam arrangement, a fixed beam arrangement is provided based on the tangent line of the planning structure geometries. Using the tangent line, the process application may i) place base set of beams on tangent line or offset from tangent line by a fixed angle; ii) place additional beams with angular offsets of 9; iii) repeat, approaching from either side of the target volume (as shown in FIG. 4). In some embodiments, the angle of each radiation delivery beam is fixed during treatment delivery. However, any of the fixed-angle radiation delivery beams can also be configured as open and collimated, as comprising several MLC segments, and / or be intensity-modulated.

[0111] Referring to FIGS. 8 and 9, there is provided an alternate example embodiment where the radiotherapy treatment modality selected by the input of the user comprises VMAT. In at least some embodiments where the radiotherapy treatment modality is VMAT, the initial set of radiation delivery geometries comprise a set of continuous radiation delivery arcs 812 (shown in FIG. 8).

[0112] In an additional embodiment, the user defined input (as part of the at least one treatment plan parameter) sets various parameters of the VMAT protocol. These parameters can include, for example, a desired number of radiation delivery arcs in the set of continuous radiation delivery arcs, and an angular offset relative to the tangential geometric reference structure for each radiation delivery arc.

[0113] In an embodiment such as shown in FIG. 8, the initial set of radiation delivery geometries for the VMAT protocol are set by defining, for each radiation delivery arc 812 of the plurality of radiation delivery arcs, an arc start angle (labelled as GANTRY START in FIG. 8) and an arc stop angle (labelled as GANTRY STOP in FIG. 8) relative to the tangent plane 200a that defines the tangential geometric reference structure 200. Additional radiation delivery arcs can be defined by varying at least one of an arc start angle or an arc stop angle (by some angular increment) across a predetermined angular search range (shown, for example, in FIG. 10).

[0114] FIG. 9 provides a flowchart of an exemplary process for defining the beam arcs for a VMAT treatment as provided in FIG. 8. In this process for defining the beam arcs, a fixed beam arc arrangement is provided based on the determined tangent line of the planning structure geometries. Using the tangent line, the process application will: set arc start / stop angles, where the angles are set according to tangent line, either at the tangent line, or offset by a fixed angle, start and stop angle can be on either side, i.e. , arc can be clockwise or counter-clockwise. The arc set in the process can be a conformal arc, i.e., with MLC conforming to projection of target volume during delivery, or can be a volumetric modulated arc, i.e., during VMAT delivery.

[0115] In some embodiments, a hybrid radiotherapy modality is selected. This hybrid radiotherapy modality can combine an arc-based radiation delivery protocol, such as the protocol shown in FIG. 8, with a fixed-beam delivery protocol, such as shown in FIG. 4. A radiation delivery arc is performed while a radiation delivery system is configured to pause at one or more preset angular positions along the arc to deliver discrete radiation beams. The discrete radiation beams delivered during the gantry pauses may be various types of beams, such as open beams or IMRT beams.Simplified Dose Calculation Model

[0116] In some embodiments, the process application is configured to perform a radiation delivery geometry update process, where the candidate radiation delivery geometries from the initial set are assessed and the updated subset of radiation deliverygeometries are selected. The simplified dose calculation model is employed to evaluate the candidate radiation delivery geometries for the selection of the updated subset of radiation delivery geometries that are included in the radiotherapy treatment pre-plan. The simplified dose calculation model may be configured to derive, for each candidate radiation delivery geometry, the respective dose estimate value to approximate a dose of radiation delivered to the non-target volume of tissue by the candidate radiation delivery geometry.

[0117] The simplified dose calculation model is less computationally complex than the full dose calculation performed by the treatment planning system and is executable using reduced computational resources and reduced execution time relative to the full dose calculation. In some example implementations, the simplified dose calculation model omits physics-based dose modeling, such as, for example, secondary radiation transport or three-dimensional scatter that is typically incorporated into the dose calculation performed by a treatment planning system.

[0118] In some embodiments, the respective dose estimate value for each candidate radiation delivery geometry is derived based on geometric characteristics of the candidate radiation delivery geometry and / or a geometric relationships between the candidate radiation delivery geometry, the target volume, and the patient anatomy (e.g., the patient surface). In this way, the simplified dose calculation model can provide an estimate of the dose delivered to the non-target tissue while avoiding a full forward dose calculation via the treatment planning system (when computing the respective dose estimate values). By decoupling the definition and screening of radiation delivery geometries from full dose calculations performed by the treatment planning system (e.g. during dose optimization and / or when performing a final forward dose calculation), the systems and methods described herein enable rapid evaluation and selection of a subset of radiation delivery geometries.

[0119] In additional embodiments, the selected subset of radiation delivery geometries can be provided to the treatment planning system (as part of the radiotherapy treatment pre-plan) or used to constrain the treatment planning system for the generation of the final radiotherapy treatment plan. Again, the use of the subset of radiation delivery geometries that are selected using the simplified dose calculation model can reduce the computational burden associated with the generating of the final radiotherapy treatment plan by the treatment planning system.

[0120] In at least some embodiments, the simplified dose calculation model is configured to derive an estimate of the respective dose estimate value for each candidate radiation delivery geometry by projecting a virtual representation of the candidate radiation delivery geometry through the contoured anatomical structures. After projecting the virtual representation of the candidate radiation delivery geometry through the contoured anatomical structures, the simplified dose calculation model quantifies an extent of overlap with the non-target volume relative to the target volume.

[0121] In an additional embodiment, the simplified dose calculation model is specifically configured to derive the respective dose estimate value based on an extent of a geometric interaction or overlap between i) a projection of a virtual representation of the candidate radiation delivery geometry through the contoured anatomical structures, and ii) the virtual representation of the non-target volume within the contoured anatomical structures. Details of example implementations of the simplified dose calculation model for a given candidate radiation delivery geometry are provided below with reference to FIGS. 12 and 13.Bolus Structure in Simplified Dose Calculation Model

[0122] As provided above, the at least one geometric planning structure can be generated automatically based on the selected body contours and the geometry of the selected target volume. In an additional embodiment, such as illustrated in FIG. 3A, the simplified dose calculation model incorporates a virtual model of a bolus dose of radiation to be delivered to the target volume. In this embodiment, the set of virtual contoured anatomical structures further includes at least one body contour defined within a portion of the non-target volume that surrounds the target volume. Additionally, at least one of the plurality of treatment plan parameters includes one or more bolus parameters that define characteristics of a bolus dose of radiation to be applied to the target volume of the patient.

[0123] The bolus parameters may be used to define a virtual bolus structure within the set of contoured anatomical structures. The virtual bolus structure represents a virtual approximation of a physical bolus and may be defined geometrically based on the target volume and one or more surrounding body contours.

[0124] For defining the virtual bolus structure, the process application may prompt a user to enter and / or select the at least one bolus parameter, such as bolus size or bolusthickness. The user-defined bolus parameter(s), together with the geometry of the virtual representation of the target volume, can then be used by the process application to define the virtual bolus structure around the target volume. In some embodiments, the process application may also provide the virtual bolus structure to the treatment planning system.

[0125] In this embodiment, the virtual bolus structure is utilized within the simplified dose calculation model as part of determining a respective dose estimate value associated with each candidate radiation delivery geometry of an initial set of candidate radiation delivery geometries. By incorporating the virtual bolus structure into the simplified dose calculation model, the dose estimate values may more accurately approximate radiation delivery conditions near the patient surface during radiotherapy treatment pre-planning.

[0126] In the example embodiment shown in FIG. 3A, the virtual bolus structure 300 is illustrated as being associated with the target volume 210. The virtual bolus structure 300 is represented as an expanded, virtual representation of the target volume 210 which has been expanded outward toward at least one body contour by an amount determined based on the bolus parameter. In this embodiment, the at least one body contour includes a first body contour 330 that surrounds the target volume 210 and a second body contour 340 that surrounds the first body contour 330. The virtual bolus structure 300 may therefore occupy a region between the target volume 210 and the surrounding body contours 330, 340 and may be used by the simplified dose calculation model to more accurately approximate dose deposition near the patient surface.

[0127] Example 5 provides another example embodiment of an algorithm for expanding the target volume to generate the virtual bolus structure.

[0128] In the specific embodiment shown in FIG. 3A, the virtual bolus structure 300 of specified thickness is defined. Various suitable configurations of the virtual bolus structure 300 may be provided. For example, the virtual bolus structure 300 may be 5 to 10 mm thick and extend 30 mm beyond the virtual representation of the target volume on the patient surface to provide adequate coverage. The extension of the target volume may be based on a geometry of the bolus, e.g., 25% of the bolus thickness. The extension of the target volume can be done to ensure that i) optimization is not performed on a very thin target volume that could present sampling issues, and ii) the final dose distribution is substantially homogeneous at the patient surface.

[0129] In some example embodiments, the system is configured to generate a virtual bolus structure that represents a sheet-type bolus, rather than a conformal bolus that closely follows a surface of the target volume.

[0130] In one such embodiment, a reference virtual bolus structure is first generated based on the target volume, where the reference virtual bolus structure establishes a spatial extent over which bolus material is to be applied. A geometric bounding region associated with the reference virtual bolus structure is then determined, and a rectangular or box-shaped virtual mask is generated based on the bounding region across a plurality of the pre-procedural image slices. A body shell of a specified thickness is further generated by expanding one of the body contours and subtracting the original body contour. The sheet-type virtual bolus structure is then defined as an intersection of the rectangular mask and the body shell, thereby producing a virtual bolus structure having substantially uniform thickness and planar extent that approximates placement of a flat bolus sheet on the patient surface.

[0131] An example process for the generation of the sheet-type, virtual bolus structure is provided in Example 4.

[0132] As further shown in FIG. 1, the process application may additionally use the body contours and target volume to generate one or more additional geometric reference structures, in addition to the virtual bolus structure, and may provide such automated planning structures to the treatment planning system.Iterative Application of Simplified Dose Calculation Model

[0133] In some embodiments, the simplified dose calculation model is applied iteratively across an initial set of candidate radiation delivery geometries to generate the updated set of radiation delivery geometries. An example flowchart for such an iterative application of the simplified dose calculation model is illustrated in FIG. 3B. In this embodiment, the simplified dose calculation model is applied sequentially to individual candidate radiation delivery geometries from the initial set of candidate radiation delivery geometries in order to generate a respective dose estimate value for each candidate radiation delivery geometry.

[0134] As shown in FIG. 3B, the method for iteratively applying the simplified dose calculation model can include a step S310 of selecting one candidate radiation delivery geometry of the initial set of candidate radiation delivery geometries, with the candidateradiation delivery geometry having an angular offset relative to the tangential geometric reference structure and a step S320 of placing the candidate radiation delivery geometry relative to the target volume and geometric reference structure. The method can further include a step S330 of determining the respective dose estimate value for the candidate radiation delivery geometry (via the dose calculation model), a step S340 of removing the candidate radiation delivery geometry from active consideration, and a step S350 of iteratively repeating steps S310 to S340 for each candidate radiation delivery geometry from the initial set of candidate radiation delivery geometries until respective dose estimate values are generated for all the candidate radiation delivery geometries. Upon completion, the subset of candidate radiation delivery geometries can then be selected based on the respective dose estimate values (at step S360).

[0135] Example methods for generating the updated set of radiation delivery geometries are provided below with reference to FIGS. 6, 7, 10, and 11.

[0136] Referring to FIGS. 6 and 7, an example embodiment is shown where the system iteratively applies the simplified dose calculation model for determining a radiation beam arrangement for an IMRT treatment pre-plan. In determining the radiation beam arrangement, the system also optimizes (based on the simplified dose calculation model) to reduce irradiation of non-target tissue by evaluating candidate beam configurations using dose estimate values generated by the simplified dose calculation model.

[0137] FIG. 6 provides a schematic diagram of an example beam arrangement for an IMRT treatment generated using the iterative optimization approach. FIG. 7 provides a flowchart illustrating an exemplary process for determining the beam arrangement shown in FIG. 6. In contrast to approaches that rely solely on beam placement defined geometrically relative to a tangential reference structure, this embodiment performs an optimization search to determine updated beam angles or configurations that minimize irradiation of non-target (healthy) tissue.

[0138] In the example embodiment illustrated in FIG. 7, inputs to the iterative optimization search may include one or more treatment plan parameters, such as a desired number of beams (N), a minimum angular separation between beams, an offset angle relative to a tangential geometric reference structure, and an angular search range. Based on these inputs, the process application initializes the optimization search by positioning a first temporary beam at a base tangential orientation or at an angular offset from the tangential geometric reference structure.

[0139] Following placement of the first temporary beam, a fast calculation of the nontarget volume of tissue that is irradiated by the temporary beam is performed using the simplified dose calculation model. An example process for performing this fast calculation is described below with reference to FIG. 12. The calculated respective dose estimate value is recorded, and the temporary beam is then removed from the beam arrangement. The process application subsequently positions a second temporary beam at an angular increment of ©search relative to the previous beam orientation. The angular increment may be positive or negative, thereby creating a beam orientation that is either more distant from the patient surface or directed further into the body. A respective dose estimate value is calculated and recorded for the second temporary beam, after which the second temporary beam is removed.

[0140] The process of temporary beam creation, evaluation, and removal continues iteratively across a selected angular range defined by M x ©search, where M represents the number of temporary beams evaluated. The angular range examined may comprise a full rotational range (e.g., 360 degrees) or may be limited to a subset of angular orientations, such as maximum angular offsets relative to the tangential geometric reference structure. Upon completion of the iterative evaluation, the process application retains a subset of beams from the optimization search. In one embodiment, N beams are retained, where N corresponds to the number of beams desired for the treatment plan, and the retained beams are those associated with the lowest recorded dose estimate value for the non-target volume of tissue.

[0141] In some embodiments, one or more additional constraints may be imposed on the selected subset of beams. For example, each retained beam may be constrained to satisfy a predetermined minimum angular separation relative to adjacent beams in the selected set. Such constraints may be applied to promote spatial diversity of beam orientations, improve plan robustness, and ensure compatibility with clinical delivery requirements.

[0142] Referring to FIGS. 10 and 11 , an example embodiment is provided where the system iteratively applies the simplified dose calculation model to determine radiation delivery arc geometries for VMAT treatment pre-plan. In this embodiment, arc start and stop angles are determined using an optimization search guided by a simplified dose calculation model, with the objective of minimizing irradiation of non-target tissue.

[0143] FIG. 10 provides a schematic diagram of an example arc arrangement for a VMAT treatment generated using the iterative optimization approach. FIG. 11 provides a flowchart illustrating an exemplary process for determining the arc arrangement shown in FIG. 10. In contrast to approaches that define arc start and stop angles solely based on a tangential geometric reference structure, this embodiment performs an optimization search to determine arc geometries that reduce the volume of irradiated non-target tissue.

[0144] In the example embodiment illustrated in FIG. 11, inputs to the iterative arc optimization may include one or more treatment plan parameters, such as a desired number of radiation delivery arcs (N), an angular offset relative to a tangential geometric reference structure, and geometric characteristics of the target volume. Based on these inputs, the process application initializes the optimization search by defining an initial radiation delivery arc having a first gantry start angle (GANTRY START0) and a first gantry stop angle (GANTRY STOP0), for example as defined relative to the tangential geometric reference structure in a manner similar to that illustrated in FIG. 8.

[0145] Following definition of the initial radiation delivery arc, a fast calculation of the respective dose estimate value corresponding to the non-target volume of tissue irradiated by the arc is performed using the simplified dose calculation model. An example process for performing this fast calculation is described below with reference to FIG. 12. The resulting dose estimate value is recorded, and the initial radiation delivery arc is then removed from consideration. The process application subsequently defines a second radiation delivery arc having gantry start and stop angles (GANTRY STARTs GANTRY STOPi) that are offset from the initial arc angles by an angular increment ©search. The dose estimate value associated with the second arc is determined and recorded, after which the second arc is removed. This process of defining, evaluating, and removing radiation delivery arcs is repeated iteratively across a predefined angular search range.

[0146] The iterative optimization search may consider a total of M radiation delivery arcs, each separated by an angular increment ©search. In some embodiments, the angular search range may comprise a full rotational range, such as 360 degrees, while in other embodiments the angular search range may be constrained based on maximum angular offsets relative to the tangential geometric reference structure. During the optimization search, the gantry range of each radiation delivery arc, defined as the difference betweenits start and stop angles, may be held fixed (for example at approximately 180 degrees) or may be varied as part of the optimization search.

[0147] Upon completion of the iterative evaluation of candidate radiation delivery arcs, the process application selects a subset of radiation delivery arcs for inclusion in the radiotherapy treatment plan as the subset of radiation delivery geometries. In one embodiment, N radiation delivery arcs are selected, where N corresponds to the number of arcs specified within the treatment plan parameters. The selected arcs may be those associated with the lowest recorded dose estimate values for the non-target volume of tissue, relative to the remaining candidate arcs.Details of Simplified Dose Calculation Model

[0148] Referring to FIGS. 12 and 13, an additional embodiment of the simplified dose calculation model is provided where the simplified dose calculation model is configured to derive the respective dose estimate value for each candidate radiation delivery geometry as a product of i) a peripheral path length through the non-target volume of tissue and ii) a projected area of the candidate radiation delivery geometry at the surface of the patient. The respective dose estimate value serves as a proxy for the actual non-target volume of tissue that is irradiated by the candidate radiation delivery geometry.

[0149] In the specific embodiment shown in FIGS. 12 and 13, the simplified dose calculation model is configured to determine the dose estimate value as a product of i) a value of a peripheral path length through the non-target volume (PLperiph) multiplied by a value of the projected area of the candidate radiation delivery geometry.

[0150] FIG. 12 provides a schematic of an exemplary method for generating a simplified dose calculation (based on the simplified dose calculation model) of the non-target volume of tissue which will be irradiated by a particular beam and beam geometry. In FIG. 12 a projection of a virtual radiation delivery beam 1200 is shown entering the surface 220 and passing through the patient body, at a central axis entry point, traversing a path through the body with a total path length (PL), intersecting the target volume 210 with a shorter path length (PLtarget), and exiting at a central axis exit point. The projected cross-sectional area of the beam through the body is indicated, and the portion of the beam path lying outside the target volume 210 corresponds to the non-target volume of tissue. Together, the beam projection, path lengths, and target intersection visually represent how a simplified dose calculation model estimates non-target irradiation based on geometric overlap rather than full dose computation.

[0151] FIG. 13 provides a flow-chart of the method for determining the peripheral path length through the non-target volume of tissue for a given radiation delivery geometry. In this embodiment, the method comprises a step S1310 of selecting a radiation delivery geometry from the plurality of initial radiation delivery geometries, and a step S1320 of determining a beam path length through the patient between a beam entry point of the and a beam exit point to define a total path length PL, a step S1330 of determining a path length through the target volume PLTarget along the beam path, a step S1340 of determining the peripheral path length as PLperiph = PL - PLTarget, and a step S1350 of calculating the volume of peripheral dose via the formula PLperiph x Abeam. This geometric estimate requires substantially less calculation time (e.g., a fraction of a second) compared to a traditional dose calculation in an existing treatment planning system.

[0152] In an additional embodiment, the simplified dose calculation model algorithm also includes beam divergence information. In the method illustrated in FIG. 12, the collimated beam is simply extruded through the volume of the patient. In a real-world application of the beam, the beam may diverge (i.e., expand in dimension with distance from the collimation). Divergence information of the beam can be used to provide a more accurate fast estimation of the irradiation of the non-target volume.

[0153] Example 2 provides further details on an alternate, example algorithm for generating a simplified dose calculation (based on the simplified dose calculation model) of the non-target volume of peripheral tissue.Selection of Subset of Radiation Delivery Geometries

[0154] As provided above, the subset of radiation delivery geometries can be selected based on respective dose estimate values generated using the simplified dose calculation model. In some embodiments, the subset of radiation delivery geometries selected for inclusion in the radiotherapy treatment pre-plan comprises radiation delivery geometries having minimized dose estimate values and / or reduced peripheral irradiation estimate values for the non-target volume relative to a remainder of the candidate radiation delivery geometries. The selected subset of radiation delivery geometries may then be used as a basis for subsequent dose optimization and forward dose calculation performed by the treatment planning system.

[0155] In some embodiments, minimization is performed by selecting the subset of candidate radiation delivery geometries that are associated with the lowest determined values of the respective dose estimate values. The dose estimate values can be rankedor otherwise compared across the initial set of candidate radiation delivery geometries, and the subset of radiation delivery geometries having the most favorable dose estimate values can be selected for inclusion in the radiotherapy treatment pre-plan.

[0156] In selecting the subset of candidate radiation delivery geometries, the system can consider and optimize one or more selection criteria.

[0157] In addition to dose-based minimization, one or more constraints can be applied when selecting the subset of radiation delivery geometries. For example, in some embodiments, the subset of radiation delivery geometries can be constrained to satisfy a predetermined minimum separation angle between adjacent radiation delivery geometries.

[0158] In some embodiments, a number of radiation delivery geometries to be included in the radiotherapy treatment pre-plan is specified via the at least one treatment plan parameters input by the user. For a given specified number of radiation delivery geometries, the system can select those candidate radiation delivery geometries that minimize estimated irradiation of the non-target volume, as represented by the respective dose estimate values generated by the simplified dose calculation model.

[0159] In some example embodiments, selection of a subset of radiation delivery geometries from an initial set of candidate radiation delivery geometries is performed using a greedy selection process based on simplified dose estimate values.

[0160] In one such embodiment, each candidate radiation delivery geometry is first assigned a dose estimate value representing an estimated irradiation of non-target tissue. The candidate radiation delivery geometries are then ranked according to their respective dose estimate values. Beginning with a candidate radiation delivery geometry having a lowest dose estimate value, radiation delivery geometries are iteratively added to the subset, provided that each newly added geometry satisfies one or more selection constraints, such as a minimum angular separation from previously selected geometries. The greedy selection process continues until a desired number of radiation delivery geometries has been selected for inclusion in a radiotherapy treatment pre-plan.

[0161] In some example embodiments, the radiotherapy treatment pre-plan includes additional optimization for dose falloff outside the target volume.

[0162] In one such embodiment, one or more additional planning structures can be generated based on the target volume and one or more body contours to support such optimization. These additional planning structures can be used by the treatment planningsystem during dose optimization and are not deliverable radiation structures. The additional planning structures can be generated automatically, manually selected, or generated using a combination of automatic and manual processes.

[0163] In an example embodiment illustrated in FIG. 3A, the one or more additional planning structures include a first tuning structure that surrounds the virtual bolus and a second tuning structure that surrounds the first tuning structure. The at least one body contour includes the first body contour around the target volume and the second body around the first body contour. The first body counter defines the first tuning structure, and the second body contour defines the second tuning structure. The first tuning structure also includes a gap relative to the virtual bolus structure and extends to a predetermined depth on a deep side of the virtual bolus structure. The second tuning structure is offset from and adjacent to the first tuning structure and extends deeper into the patient than the first tuning structure to limit intermediate-to-low irradiation outside the target volume.

[0164] In an additional embodiment, the system is configured to constrain the dose optimization performed by the treatment planning system using at least the first tuning structure and the second tuning structure as optimization constraint structures. The first tuning structure can be provided to ensure conformity of the high dose to the target volume. Typically, the first tuning structure extends around the virtual bolus structure and will involve a gap between the target volume (e.g., of about 2 or 3 mm) to avoid compromising dose coverage of the target volume. The first tuning structure extends a specified thickness at the deep aspect of the target volume (e.g., about 1 to 2 cm). The second tuning structure can be provided to limit intermediate to low doses to normal tissues. The proximal (to skin) aspect of this tuning structure will typically abut the first tuning structure and will extend at the deep aspect a total of, e.g., 5 to 10 cm.

[0165] By incorporating these tuning structures into the dose optimization process, the treatment planning system can promote conformity of a high-dose region to the target volume while controlling dose falloff within the non-target volume of tissue.

[0166] Once the updated radiation delivery geometries have been generated, the process application may provide the updated radiation delivery geometry to the treatment planning system and prompt the treatment planning system to perform a dose optimization process. As shown in FIG. 1A, the process application prompts the user to input dose optimization parameters and sends the dose optimization parameters on to the treatment planning system. The process application can then call an optimizationroutine from the treatment planning system, while also providing the user-defined treatment details (e.g., IMRT or VMAT) and the radiotherapy treatment pre-plan. The treatment planning system can then perform dose calculation and dose optimization.

[0167] In accordance with the present disclosure, the radiotherapy treatment plan generated by the treatment planning system based on the radiotherapy treatment pre-plan may be provided to a radiation delivery system for delivery of radiation to the patient. In some embodiments, the radiotherapy treatment plan includes machine-readable treatment parameters that define radiation delivery geometries, beam or arc parameters, multi-leaf collimator (MLC) configurations, dose rates, and timing information. The treatment planning system may export or otherwise make available the radiotherapy treatment plan in a format compatible with the radiation delivery system.

[0168] The radiation delivery system may be any suitable system configured to deliver external beam radiation therapy. By way of example and not limitation, the radiation delivery system may include a linear accelerator (linac) configured for IMRT, VMAT, or three-dimensional conformal radiation therapy (3D-CRT), a C-arm-based linear accelerator, a ring-based delivery systems, other hybrid delivery systems combining arc-based and fixed-beam delivery, or other gantry-mounted radiation delivery platforms.

[0169] In some example embodiments, the system is configured to automatically detect and correct undeliverable radiation delivery arcs arising from physical constraints of a radiation delivery system (e.g., a Gantry crossing).

[0170] In one such embodiment, arc start and stop angles determined during radiotherapy treatment pre-planning are analyzed to determine whether a continuous arc would require a gantry to cross a rotational boundary that is not physically traversable during delivery. When such a condition is detected, the system generates one or more alternative arc configurations corresponding to different rotational directions or adjusted angular endpoints. The alternative arc configurations may be evaluated using geometric or physical criteria, such as average source-to-surface distance.

[0171] Example 6 provides another example embodiment of an algorithm to detect and correct the undeliverable radiation delivery arcs.

[0172] As provided above, the system 100 includes processing hardware with the at least one processor and memory. Referring to FIG. 14, an example schematic of the processing hardware for implementing the one or more process applications andinterfacing with the existing treatment planning system 140, is shown. The processing hardware 500 may include a processor 510, a memory 515, a system bus 505, one or more input / output devices 520, and a plurality of optional additional devices such as communications interface 535, display 560, external storage 530, and data acquisition interface 540. In one example implementation, the display 560 may be employed to provide a user interface facilitating input and displaying output of the treatment plan, optionally as an overlay window / application on another user interface of the treatment planning system 570. As shown in FIG. 14, the display and / or the treatment planning system 570 may be directly integrated with the control and processing hardware 500 that implements the present example application program, as shown at 590, or may be provided as an external device or system that is interfaced with the control and processing hardware 500.

[0173] The systems and methods described herein can be implemented via processor 510 and / or memory 515. As shown in FIG. 14, executable instructions represented as reference geometry analysis module 580 are processed by control and processing hardware 500 to generate a tangent or other suitable geometric reference based on input volumetric image data, and modules 582 and 584 include executable instructions for performing the beam planning operations and peripheral dose volume calculations described above, respectively. Such executable instructions may be stored, for example, in the memory 515 and / or other internal storage.

[0174] The methods described herein can be partially implemented via hardware logic in processor 510 and partially using the instructions stored in memory 515. Some embodiments may be implemented using processor 510 without additional instructions stored in memory 515. Some embodiments are implemented using the instructions stored in memory 515 for execution by one or more microprocessors. Thus, the disclosure is not limited to a specific configuration of hardware and / or software.

[0175] It is to be understood that the example system shown in the FIG. 14 is not intended to be limited to the components that may be employed in a given implementation. For example, the system may include one or more additional processors. Furthermore, one or more components of control and processing hardware 500 may be provided as an external component that is interfaced to a processing device. Furthermore, although the bus 505 is depicted as a single connection between all of the components, it will be appreciated that the bus 505 may represent one or more circuits,devices or communication channels which link two or more of the components. For example, the bus 505 may include a motherboard. The control and processing hardware 500 may include many more or less components than those shown.

[0176] Some aspects of the present disclosure can be embodied, at least in part, in software, which, when executed on a computing system, transforms an otherwise generic computing system into a specialty-purpose computing system that is capable of performing the methods disclosed herein, or variations thereof. That is, the techniques can be carried out in a computer system or other data processing system in response to its processor, such as a microprocessor, executing sequences of instructions contained in a memory, such as ROM, volatile RAM, non-volatile memory, cache, magnetic and optical disks, or a remote storage device. Further, the instructions can be downloaded into a computing device over a data network in a form of compiled and linked version. Alternatively, the logic to perform the processes as discussed above could be implemented in additional computer and / or machine-readable media, such as discrete hardware components as large-scale integrated circuits (LSI's), application-specific integrated circuits (ASIC's), or firmware such as electrically erasable programmable readonly memory (EEPROM's) and field-programmable gate arrays (FPGAs).

[0177] A computer readable storage medium can be used to store software and data which when executed by a data processing system causes the system to perform various methods. The executable software and data may be stored in various places including for example ROM, volatile RAM, nonvolatile memory and / or cache. Portions of this software and / or data may be stored in any one of these storage devices. As used herein, the phrases “computer readable material” and “computer readable storage medium” refers to all computer-readable media, except for a transitory propagating signal perse.

[0178] While many of the example embodiments disclosed herein pertain to the preplanning of beam configurations and radiation treatment parameters for radiation therapy of skin cancers, it will be understood that the example systems and methods disclosed herein may be adapted for radiation treatment of a wide variety of pathologies. Indeed, as noted above, the process application as described herein can be applied for planning radiation treatment of other types of radiation-treatable diseases. The planning structure geometries can be selected and defined based on the type of disease to be treated.

[0179] For example, non-limiting examples of other clinical applications include: i) prostate cancer: using an IMRT beam arrangement that, following dose optimization,maximizes the dose fall-off from the prostate to the rectal wall. In this case the geometric reference would be, e.g., a plane between the posterior aspect of the prostate and the anterior rectal wall; ii) breast cancer: breast is typically treated with a tangential approach, similar to skin lesions. Here the reference line would be a tangent to the chest wall; and iii) lung cancer: here a beam or arc arrangement would be based on the laterality of the target volume. For example, in treating a peripheral lesion on the patients left, a beam or arc arrangement would be biased to the ipsilateral side.Enumerated Embodiments

[0180] Embodiment 1. A system for generating a radiotherapy treatment pre-plan for the delivery of a dose of radiation to a target volume of a patient, the system comprising:processing hardware comprising at least one processor and associated memory, the memory storing instructions that, when executed by the at least one processor, cause the system to perform operations that include:obtaining a set of virtual, contoured anatomical structures, at least some of the contoured anatomical structures defining the target volume of the patient, a non-target volume of the patient, and a patient surface;receiving an input from the user that defines a radiotherapy treatment modality and at least one treatment plan parameter for the radiotherapy treatment pre-plan;defining, based on the target volume, a tangential geometric reference structure which approximates a tangent to at least part of a surface-facing portion of a contoured anatomical structure that defines the target volume, the surfacefacing portion being oriented toward the patient surface;employing the selected radiotherapy treatment modality, the at least one treatment plan parameter, and the at least one tangential geometric reference structure to determine an initial set of candidate radiation delivery geometries that are defined relative to the tangential geometric reference structure;employing a simplified dose calculation model to estimate, for each candidate radiation delivery geometry of the initial set of candidate radiation delivery geometries, a respective dose estimate value that corresponds to an irradiation of the non-target volume of the patient, wherein the respective dose estimate value is derived based on an extent of a geometric interaction or overlapbetween i) a virtual projection of the candidate radiation delivery geometry through the contoured anatomical structures, and ii) the non-target volume within the contoured anatomical structures; andselecting a subset of candidate radiation delivery geometries from the initial set of candidate radiation delivery geometries for inclusion in an updated radiotherapy treatment pre-plan based on a reduction of dose delivered to the non-target volume, as represented by the respective dose estimate of each candidate radiation delivery geometry.

[0181] Embodiment 2. The system of embodiment 1, wherein the subset of candidate radiation delivery geometries have reduced respective dose estimate values for the non-target volume relative to the respective dose estimate values for a remainder of the initial set of candidate radiation delivery geometries.

[0182] Embodiment s. The system of embodiment 1, wherein the subset of candidate radiation delivery geometries selected from the initial set delivery geometries are associated with lowest determined values of the respective dose estimate values.

[0183] Embodiment 4. The system of embodiment 1, wherein obtaining the set of virtual, contoured anatomical structures comprises:receiving a pre-procedural image dataset that includes pre-procedural image data of the set of virtual, contoured anatomical structures;receiving an input from a user that identifies the target volume within the set of virtual, contoured anatomical structures.

[0184] Embodiment s. The system of embodiment 1, wherein each candidate radiation geometry of the initial set of candidate radiation geometries is defined at the tangential geometric reference structure or at one or more angular offsets relative to the tangential geometric reference structure.

[0185] Embodiments. The system of embodiment 5, wherein the initial set of candidate radiation delivery geometries defines at least one of i) a set of radiation delivery beams and ii) a set of continuous radiation delivery arcs, as per the selected radiotherapy treatment modality.

[0186] Embodiment ?. The system of embodiment 1, wherein the subset of radiation delivery geometries selected for inclusion in the updated radiotherapy treatment pre-plan have minimized dose estimate value associated therewith.

[0187] Embodiment 8. The system of any one of embodiments 1 to 7, wherein the instructions stored on the memory further cause the system to perform operations that include:providing the updated radiotherapy treatment plan and the at least one treatment plan parameter to a treatment planning subsystem for performing forward dose calculation and dose normalization.

[0188] Embodiment 9. The system of embodiment 1, wherein the surface facing portion of the target volume is proximate the patient surface; andwherein the at least one tangential geometric reference structure is defined through a point on the patient surface and has an orientation determined based on a local surface geometry of a part of the surface facing portion of the target volume that is closest to the surface of the patient.

[0189] Embodiment 10. The system of embodiment 4, wherein the pre-procedural image data includes a plurality of image slices of the patient that encompass the target volume.

[0190] Embodiment 11. The system of embodiment 10, wherein the at least one tangential geometric reference structure comprises a tangent line or tangent plane that is determined as an aggregate tangent for the target volume across a subset of the plurality of image slices which span a superior-inferior extent of the target volume.

[0191] Embodiment 12. The system of embodiment 10, wherein the at least one tangential geometric reference structure comprises a tangent line or tangent plane determined on one image slice that is defined through a superior-inferior midpoint of the target volume.

[0192] Embodiment 13. The system of embodiment 10, wherein the virtual, contoured anatomical structures define a three-dimensional representation of the target volume of the patient; andwherein defining the tangential geometric reference structure comprises reducing the three-dimensional representation of the target volume to a two-dimensional analysis on a representative image slice of the plurality of image slices that include the target volume.

[0193] Embodiment 14. The system of embodiment 13, wherein defining the tangential geometric reference structure further comprises:

[0194] extracting a two-dimensional point set corresponding to a contour of the target volume on the representative image slice; and computing a tangent angle of the tangential geometric reference structure based on geometric extrema of the two-dimensional point set.

[0195] Embodiment 15. The system of embodiment 1, wherein the set of virtual, contoured anatomical structures further comprises at least one body contour defined in a portion of the non-target volume that surrounds the target volume.

[0196] Embodiment 16. The system of embodiment 15, wherein the at least one treatment plan parameter comprises a plurality of treatment plan parameters; and

[0197] wherein at least one of the plurality of treatment plan parameters includes a bolus parameters for a bolus dose of the radiation to be applied to the target volume of the patient.

[0198] Embodiment 17. The system of embodiment 16, wherein the instructions stored on the memory further cause the system to perform operations that include:defining a virtual bolus structure associated with the target volume based on the at least one bolus parameter;wherein the virtual bolus structure is utilized within the simplified dose calculation model to determine the respective dose estimate value for with each candidate radiation delivery geometry of the initial set of candidate radiation delivery geometries.

[0199] Embodiment 18. The system of embodiment 17, wherein defining the virtual bolus structure comprises expanding a virtual representation of the target volume outward toward the at least one body contour by an amount that is established from the at least one bolus parameter.

[0200] Embodiment 19. The system of embodiment 18, wherein the instructions stored on the memory further cause the system to: automatically generate one or more additional planning structures based on the target volume and the at least one body contour, wherein the one or more additional planning structures are used by the treatment planning subsystem during dose optimization and are not deliverable radiation structures.

[0201] Embodiment 20. The system of embodiment 19, wherein the one or more additional planning structures comprise a first tuning structure that surrounds the virtualbolus, the first tuning structure including a gap relative to the virtual bolus structure and extending to a predetermined depth on a deep side of the virtual bolus structure.

[0202] Embodiment 21. The system of embodiment 20, wherein the one or more additional planning structures comprise a second tuning structure that is offset from, and adjacent to, the first tuning structure, and extends deeper into the patient than the first tuning structure to limit intermediate-to-low irradiation outside the target volume.

[0203] Embodiment 22. The system of embodiment 21, wherein the system is configured to constrain the dose optimization of the treatment planning subsystem using at least the first tuning structure and the second tuning structure as optimization constraint structures to thereby promote conformity of a high-dose region to the target volume and control dose falloff in the non-target volume of tissue.

[0204] Embodiment 23. The system of embodiment 21 or 22, wherein the at least one body contour includes a first body contour that immediately surrounds the target volume and a second body contour that surrounds the first body contour; and wherein the first body counter defines the first tuning structure, and the second body contour defines the second tuning structure.

[0205] Embodiment 24. The system of embodiment 6, wherein the radiotherapy treatment modality selected by the input of the user comprises intensity-modulated radiation therapy (IMRT);wherein the initial set of radiation delivery geometries comprises a set of fixed-angle radiation delivery beams; andwherein the at least one treatment plan parameter includes a desired number fixed-angle radiation delivery beams, a minimum separation angle between adjacent fixed-angle radiation delivery beams of the set of fixed-angle radiation delivery beams, and an angular offset relative to the tangential geometric reference structure.

[0206] Embodiment 25. The system of embodiment 24, wherein generating the initial set of radiation delivery geometries comprises placing the set of fixed-angle radiation delivery beams at angular positions defined relative to the tangential geometric reference structure.

[0207] Embodiment 26. The system of embodiment 25, wherein placing the set of fixed-angle radiation delivery beams at angular positions comprises:placing a first pair of fixed-angle radiation delivery beams at angular positions defined by the angular offset relative to the tangential geometric reference structure,wherein the first pair of fixed-angle radiation delivery beams are positioned on opposing sides of the tangential geometric reference structure; andplacing a remainder of the set of fixed-angle radiation delivery beams at angular positions that are spaced from one another and from the first pair of fixed-angle radiation delivery beams by at least the minimum separation angle.

[0208] Embodiment 27. The system of embodiment 6, wherein the radiotherapy treatment modality selected by the input of the user comprises volumetric modulated arc therapy (VMAT);wherein the initial set of radiation delivery geometries comprise the set of continuous radiation delivery arcs; andwherein the at least one treatment plan parameter includes a desired number of radiation delivery arcs in the set of continuous radiation delivery arcs and an angular offset relative to the tangential geometric reference structure.

[0209] Embodiment 28. The system of embodiment 27, wherein generating the initial set of radiation delivery geometries comprises defining, for each radiation delivery arc of the plurality of radiation delivery arcs, an arc start angle and an arc stop angle relative to the tangential geometric reference structure.

[0210] Embodiment 29. The system of embodiment 1, wherein employing the simplified dose calculation model comprises:i) selecting one candidate radiation delivery geometry of the initial set of candidate radiation delivery geometries, the candidate radiation delivery geometry having an angular offset relative to the tangential geometric reference structure;ii) determining the respective dose estimate value for the candidate radiation delivery geometry;iii) removing the candidate radiation delivery geometry; andiv) iteratively repeating steps i) to iii) for each candidate radiation delivery geometry from the initial set of candidate radiation delivery geometries for generating the respective dose estimate values for all the candidate radiation delivery geometries;wherein the subset of candidate radiation delivery geometries are selected based on the respective dose estimate values.

[0211] Embodiment 30. The system of embodiment 29, wherein the candidate radiation delivery geometries comprise radiation delivery arcs, and wherein applying the radiation geometry optimization process comprises generating the radiation delivery arcsby varying at least one of an arc start angle or an arc stop angle by an angular increment across a predetermined search range.

[0212] Embodiment 31. The system of embodiment 29, wherein the subset of candidate radiation delivery geometries selected from the initial set delivery geometries are associated with lowest determined values of the respective dose estimate values.

[0213] Embodiment 32. The system of embodiment 29, wherein the subset of radiation delivery geometries are constrained to satisfy a predetermined, minimum separation angle between adjacent radiation delivery geometries of the set of radiation delivery geometries.

[0214] Embodiment 33. The system of embodiment 1, wherein the simplified dose calculation model derives the respective dose estimate value for each candidate radiation delivery geometry as a product of i) a peripheral path length through the non-target volume of tissue and ii) a projected area of the candidate radiation delivery geometry at the surface of the patient.

[0215] Embodiment 34. The system of embodiment 33, wherein the peripheral path length through the non-target volume of tissue is computed by:determining a beam path length through the patient between a beam entry point of the and a beam exit point to define a total path length PL;determining a path length through the target volume PLTarget along the beam path; anddetermining the peripheral path length as PLperiph = PL - PLTarget.

[0216] Embodiment 35. The system of embodiment 34, wherein the dose estimate value comprises PLperiph multiplied by the projected area.

[0217] Embodiment 36. The system of embodiment 1, wherein the instructions stored on the memory further cause the system to perform operations that include:generating machine-readable radiotherapy treatment plan data that configure a radiation delivery system to deliver radiation according to the selected subset of radiation delivery geometries.

[0218] Embodiment 37. The system of embodiment 1, wherein the instructions stored on the memory further cause the system to perform operations that include:applying the subset of radiation delivery geometries to constrain forward dose calculation and dose normalization in the treatment planning subsystem during the generation of the radiotherapy treatment plan data.

[0219] Embodiment 38. The system of embodiment 1, further comprises a user interface that includes a display and that is operably connected to the processing hardware;wherein the instructions stored on the memory further cause the system to perform operations that include generating a visual rendering of at least part of the preprocedural image data and presenting the at least part of the pre-procedural image data to the user via the display.

[0220] Embodiment 39. A system for generating a radiotherapy treatment pre-plan for the delivery of a dose of radiation to a target volume of a patient, the system comprising:processing hardware comprising at least one processor and associated memory, the memory storing instructions that, when executed by the at least one processor, cause the system to perform operations that include:obtaining a set of virtual, contoured anatomical structures, at least some of the contoured anatomical structures defining the target volume of the patient, a nontarget volume of the patient, and a patient surface;receiving an input from the user that defines a radiotherapy treatment modality and at least one treatment plan parameter for the radiotherapy treatment preplan;defining, based on the target volume, at least one geometric reference structure associated with the target volume;employing the selected radiotherapy treatment modality, the at least one treatment plan parameter, and the at least one geometric reference structure to determine an initial set of candidate radiation delivery geometries defined relative to the geometric reference structure;employing a simplified dose calculation model to estimate, for each candidate radiation delivery geometry of the initial set of candidate radiation delivery geometries, a respective dose estimate value that corresponds to an irradiation of the non-target volume of the patient;selecting a subset of candidate radiation delivery geometries from the initial set of candidate radiation delivery geometries based on a reduction of dose delivered to the non-target volume; andproviding the selected subset of radiation delivery geometries to a treatment planning system that is configured to generate a radiotherapy treatment plan by performing a full dose calculation based on the selected subset of radiation delivery geometries;wherein the simplified dose calculation model that is less computationally complex than the full dose calculation performed by the treatment planning system and is executable using reduced computational resources and reduced execution time relative to the full dose calculation.

[0221] Embodiment 40. A system for generating a radiotherapy treatment pre-plan for the delivery of a dose of radiation to a target volume of a patient, the system comprising:processing hardware comprising at least one processor and associated memory, the memory storing instructions that, when executed by the at least one processor, cause the system to perform operations that include:obtaining a set of virtual, contoured anatomical structures, at least some of the contoured anatomical structures defining the target volume of the patient, a nontarget volume of the patient, and a patient surface;receiving an input from the user that defines a radiotherapy treatment modality and at least one treatment plan parameter for the radiotherapy treatment preplan;defining, based on the target volume, at least one geometric reference structure associated with the target volume;employing the selected radiotherapy treatment modality, the at least one treatment plan parameter, and the at least one geometric reference structure to determine an initial set of candidate radiation delivery geometries that are defined relative to the geometric reference structure;employing a simplified dose calculation model to estimate, for each candidate radiation delivery geometry of the initial set of candidate radiation delivery geometries, a respective dose estimate value that corresponds to an irradiation of the non-target volume of the patient;selecting a subset of candidate radiation delivery geometries from the initial set of candidate radiation delivery geometries for inclusion in an updated radiotherapytreatment pre-plan based on a reduction of dose delivered to the non-target volume; andproviding the selected subset of radiation delivery geometries to a treatment planning system for the generation of a radiotherapy treatment plan;wherein the selecting of the subset of candidate radiation delivery geometries is performed without prompting the treatment planning system to perform iterative dose optimization.EXAMPLES

[0222] The following examples are presented to enable those skilled in the art to understand and to practice embodiments of the present disclosure. They should not be considered as a limitation on the scope of the disclosure, but merely as being illustrative and representative thereof.Example 1 - Details steps for example tangent angle algorithm

[0223] The following examples provides details of the example tangent angle algorithm for determining the at least one tangent angle, as shown in FIG. 2B. The tangent angle algorithm begins by computing the axial midpoint of the tangent volume from a mesh geometry of the tangent volume. All mesh vertex positions are retrieved, and the minimum and maximum Z coordinates are determined. The midpoint is computed as:Zmid=Zmin+(Zmax—Zmin) I 2Starting from Zmid and searching upward through the ordered Z-coordinate list, the algorithm selects the first Z-plane that contains more than a minimum threshold of contour points (default: 4 points). This threshold ensures that the selected slice has sufficient geometric information for a meaningful tangent calculation, avoiding degenerate slices near the superior or inferior extent of the target where only a few contour points may be present.

[0224] At the selected Z-plane, the X and Y coordinates of all mesh vertices are extracted into separate lists. The geometric extents are computed:Xdim=Xmax—Xminydim=ymax—yminThese extents define the width and height of the target volume cross-section in the axial plane and determine which calculation branch will be used.

[0225] Next, the algorithm selects one of two calculation methods based on the aspect ratio of the target volume cross-section. Where the target volume is a wide target (xdim > ydim), the algorithm identifies the contour points at the leftmost (xmin) and rightmost (Xmax) positions, and retrieves their corresponding Y coordinates (yxmin and yxmax). The tangent angle is computed from the Y-coordinate difference across the full X span. Where the target volume is a tall target (ydim > Xdim), the algorithm identifies the contour points at the bottommost (ymin) and topmost (ymax) positions, and retrieves their corresponding X coordinates (xymin and xymax). The tangent angle is computed from the Y span across the X-coordinate difference. This adaptive branching ensures that the tangent angle calculation uses the most geometrically meaningful reference points for the given target shape. For a wide target, the lateral extrema provide the strongest angular signal; for a tall target, the superior-inferior extrema are more informative.

[0226] Lastly, the tangent angle is computed using the two-argument arctangent function:For a wide target (xdim > ydim):0=arctan[(yxmin - yx max ) / (x max Xmin)] X (180° I TT)For a tall target (ydim > Xdim):0=arctan[(ymax—ymin) I (Xymin—Xymax)]x(180° I TT)The result is a signed angle in degrees representing the tilt of a cross-section of the target volume relative to the cardinal axes.

[0227] The complete mathematical formulation for the tangent angle calculation is presented below. Let P = {(x, yi) | i = 1..n} be the set of contour points on the representative axial slice.

[0228] Define:Xmin = min{ Xi }, xmax = max{ x }ymin = min{ yi }, ymax = max{ yi }Xdim=Xmax ~ Xmin, ydim= / max—ymin

[0229] Case 1 : Xdim > ydim (wide target).Let (xmin, yxmin) be the point where x = Xmin, and (Xmax , yxmax) be the point where X=Xmax . Then:Q=arctan[(yxmin ~ yxmax) / (Xmax ~ Xmin)]x(180 / TT)

[0230] Case 2: ydim > Xdim (tall target)Let (xymin, ymin) be the point where y = ymin, and (xymax, ymax ) be the point where y=ymax . Then:9=3TCtdn[(ymax ~ Ymin) / (Xymin ~ Xymax)]x(180 / TT)

[0231] Gantry angle transformation:For HFS or HFP: a1= 27O°- 0, a2= 90° - 0For FFS or FFP: a7= 270°+ 0 a2= 90°+ 9

[0232] This example implementation relies on several assumptions and constraints. First, the 3D target volume geometry is reduced to a 2D cross-section at a single representative axial slice. This assumes that a tangent direction does not vary significantly along the cranial-caudal axis. In addition, the tangent angle algorithm computes angles in the transverse (axial) plane only. Non-coplanar beam arrangements are not considered. At least 4 contour points must be present on the selected slice for the algorithm to proceed. Slices with fewer points are skipped to avoid degenerate geometry. The algorithm also has a preset number of usable orientations. Four standard patient orientations are supported (HFS, HFP, FFS, FFP). Furthermore, when multiple contour points share the same extreme coordinate value, the first match encountered is used. This is adequate for typical convex or mildly concave targets but may be suboptimal for highly irregular contours. Lastly, the algorithm operates on mesh vertices rather than analytical contour representations. The vertex density depends on the Eclipse mesh generation parameters and may not perfectly represent the underlying structure contours.Example 2 - Peripheral dose calculation

[0233] This example provides another example embodiment of calculations within the simplified dose calculation model for determining a respective dose estimate value for a given candidate radiation delivery geometry. Again, the respective dose estimate value is a computationally efficient proxy for the dose deposited in tissue beyond the target volume by a beam at a given angle without a full dose calculation.

[0234] The respective dose estimate value is based on two geometric quantities: MLC projected area (A) and patient separation (S). In this calculation, a test beam is created at the candidate angle with the MLC automatically fitted to the two-dimensional contour for the target volume. An open field area is computed by summing the gap between opposing leaf pairs across all leaf positions. This represents the projected cross-section of the target as seen from the beam's-eye view. For the patient separation (S), athickness of patient tissue traversed by the beam is determined by creating an opposing beam (180° rotation) and measuring the SSD at both the original and opposing angles. The separation is calculated as:S = SADtotai ~ SSDb earn SS D oppositewhere SADtotai is twice the source-to-axis distance (2 x 1000 mm = 2000 mm for standard linacs). The peripheral dose estimate is then:D_peripheral ~A * S

[0235] The physical intuition is that a beam traversing a longer path through the patient (larger S) with a larger field opening (larger A) will deposit more dose outside the target. While this is not a dose calculation in the physics sense, it provides a ranking metric that correctly identifies beam angles with lower unwanted dose to normal tissue. This metric also enables evaluation of several hundred candidate angles in seconds, where a full dose calculation at each candidate would require minutes per angle.Example 3 - Body Structure Preservation and Irregular Bolus Creation

[0236] This example provides an algorithm-based method for body structure preservation and irregular bolus creation.

[0237] When a bolus is added to a plan, the patient body contour must be expanded to include the bolus volume. The Eclipse API does not permit assignment of the BODY structure type to a newly created structure. DermaPlanRT addresses this constraint through the following procedure:1. The original user body structure is backed up into a separate control structure (dpOriginalBody) that preserves the unmodified body contour for subsequent runs. 2. The new body volume is computed as the Boolean union of the original body and the bolus structure.3. If a previously modified body structure (dpBodyWithBolus) exists from a prior run, its segment volume is overwritten in place, preserving the BODY type assignment.4. If no prior modified body exists, the original body structure's segment volume is replaced with the new volume and its identifier is updated, preserving the BODY type on the same structure object.

[0238] This approach ensures the treatment planning system continues to recognize the structure as the patient body for dose calculation purposes, which requires the BODY (or EXTERNAL) DICOM type designation.

[0239] An irregular bolus can be created using three volumetric Boolean operations that produce a structure conforming to the target volume shape on the patient surface: 1. A margin expansion of the target volume by the bolus surround distance creates an envelope around the target.2. The patient body is expanded outward by the bolus thickness, and the original body is subtracted to create a body shell (ring) of the specified thickness.3. The target volume envelope is intersected (Boolean AND) with the body shell. The resulting bolus volume is the irregular bolus. It follows the contour of both the target volume and the patient surface and is represented based on the following:V bolus=(PTV (P surround) D [(Body (p tbolus) \ Body]where ® denotes morphological dilation (margin expansion), n denotes intersection, and \ denotes set subtraction. The parameters Tsurround and tboius are the bolus surround and thickness in millimeters, respectively. Both are capped at 50 mm, which is the maximum margin supported by the Eclipse API.Example 4 - Sheet Bolus via Bounding Box Algorithm

[0240] This example provides a bound-box-based algorithm for the generation of a sheet virtual bolus structure. Clinical practice frequently uses flat sheet bolus material (e.g., SuperFlab) rather than custom-molded irregular bolus. This algorithm includes the following steps:1. An irregular reference bolus (range bolus) is first created using the method described in Section 4.2 to establish the spatial extent of the bolus region.2. The axis-aligned bounding box of the reference bolus mesh is extracted, yielding the minimum and maximum coordinates in X, Y, and Z.3. The Z-coordinate bounds are converted to CT slice indices. A rectangular mask structure is created by drawing a closed rectangle with the X-Y bounds of the bounding box on each axial slice within the Z range.4. A body shell of the specified thickness is computed by subtracting the original body from a dilated body.5. The sheet bolus is the Boolean intersection of the body shell and the rectangular box mask. This produces a uniform -thickness layer that covers the full rectangular extent of the PTV region on the patient surface, faithfully representing the clinical placement of a flat bolus sheet.Vsheet = [(Body t) | Body] n BoxMask(BoundingBox(VrangeBoius))The reference bolus and box mask are temporary structures that are removed after computation, leaving only the final sheet bolus.Example 5 - Extending Target Volume

[0241] This example provides an algorithm for extending the target volume to form the virtual bolus structure.

[0242] A challenge in bolus-based skin planning is achieving adequate dose coverage at the interface between the patient body surface and the bolus. If the optimizer targets only the original PTV, dose may fall off sharply at this boundary. This example embodiment for generating the virtual bolus structure creates an expanded virtual target volume that bridges this interface. The algorithm for extending the target volume is as follows:1. The PTV is expanded asymmetrically in the lateral directions (X and Y) by one-quarter of the bolus thickness. No expansion is applied in the Z (cranial-caudal) direction.2. The original body contour (without bolus) is subtracted from this expansion, isolating the portion that lies within the bolus region but outside the native body.3. This bolus-region portion is merged (Boolean OR) with the original PTV to create the extended target.4. The result is trimmed by intersecting with the bolus-augmented body minus a small safety margin (2 mm), preventing the extended PTV from protruding beyond the outer bolus surface.PTVext = [PTV u ((PTV xy t / 4) | Bodyong)] n (Bodyboius 32mm)The extended PTV is assigned PTV structure type and serves as the primary optimization target for VMAT and IMRT plans, while the original PTV is used for normalization and dose reporting.Example 6 - Gantry 180° Crossing Detection and Correction

[0243] This example provides an algorithm for the quadrant-based detection of undeliverable arcs that would cross the 180° gantry limit, with SSD-based arc direction selection to determine the correct treatment side. Generally, a linear accelerator gantry can rotate from 0° clockwise to 180° or from 0° counterclockwise to 180°, but it cannot continuously pass through the 180° position during arc delivery. When the computedtangent angles (e.g., those computed by the tangent angle algorithm) span this boundary, the resulting arc would be physically undeliverable.

[0244] This example algorithm for the quadrant-based detection of undeliverable arcs detects this condition using quadrant analysis. The gantry angle space is divided into four quadrants as viewed from the foot of the treatment table:Quadrant Angle Range PositionQ1 0° - 90° Right-anteriorQ2 90° - 180° Right-posteriorQ3 180° -270° Left-posteriorQ4 270° - 360° Left-anterior

[0245] A 180° crossing issue is determined to exist when the start and stop angles lie in quadrants Q1 and Q3, or in Q2 and Q4. When detected, the correction algorithm proceeds as follows:1. A clockwise (CW) test arc is created between the start and stop angles. Its average SSD is recorded.2. A counterclockwise (CCW) test arc is attempted. If it fails (because it would cross 180°), the problematic angle is identified and moved to a safe position (179° or 181 ° as appropriate). The CCW arc is then recreated with the corrected angle.3. The average SSD of both arcs is compared.

[0246] The arc with the larger average SSD is the one approaching from outside the body (the clinically correct direction). If the CCW arc is selected, the start and stop angles are reversed to produce an equivalent CW arc specification, preserving the safe angle correction.

[0247] The specific embodiments described above have been shown by way of example, and it should be understood that these embodiments may be susceptible to various modifications and alternative forms. It should be further understood that the claims are not intended to be limited to the particular forms disclosed, but rather to cover all modifications, equivalents, and alternatives falling within the spirit and scope of this disclosure.

Claims

1. CLAIMS1. A system for generating a radiotherapy treatment pre-plan for the delivery of a dose of radiation to a target volume of a patient, the system comprising:processing hardware comprising at least one processor and associated memory, the memory storing instructions that, when executed by the at least one processor, cause the system to perform operations that include:obtaining a set of virtual, contoured anatomical structures, at least some of the contoured anatomical structures defining the target volume of the patient, a non-target volume of the patient, and a patient surface;receiving an input from the user that defines a radiotherapy treatment modality and at least one treatment plan parameter for the radiotherapy treatment pre-plan;defining, based on the target volume, a tangential geometric reference structure which approximates a tangent to at least part of a surface-facing portion of a contoured anatomical structure that defines the target volume, the surface-facing portion being oriented toward the patient surface;employing the selected radiotherapy treatment modality, the at least one treatment plan parameter, and the at least one tangential geometric reference structure to determine an initial set of candidate radiation delivery geometries that are defined relative to the tangential geometric reference structure;employing a simplified dose calculation model to estimate, for each candidate radiation delivery geometry of the initial set of candidate radiation delivery geometries, a respective dose estimate value that corresponds to an irradiation of the non-target volume of the patient, wherein the respective dose estimate value is derived based on an extent of a geometric interaction or overlap between i) a virtual projection of the candidate radiation delivery geometry through the contoured anatomical structures, and ii) the non-target volume within the contoured anatomical structures; andselecting a subset of candidate radiation delivery geometries from the initial set of candidate radiation delivery geometries for inclusion in an updated radiotherapy treatment pre-plan based on a reduction of dose delivered to the non-target volume, as represented by the respective dose estimate of each candidate radiation delivery geometry.

2. The system of claim 1, wherein the subset of candidate radiation delivery geometries have reduced respective dose estimate values for the non-target volume relative to the respective dose estimate values for a remainder of the initial set of candidate radiation delivery geometries.

3. The system of claim 1, wherein the subset of candidate radiation delivery geometries selected from the initial set delivery geometries are associated with lowest determined values of the respective dose estimate values.

4. The system of claim 1, wherein obtaining the set of virtual, contoured anatomical structures comprises:receiving a pre-procedural image dataset that includes pre-procedural image data of the set of virtual, contoured anatomical structures;receiving an input from a user that identifies the target volume within the set of virtual, contoured anatomical structures.

5. The system of claim 1 , wherein each candidate radiation geometry of the initial set of candidate radiation geometries is defined at the tangential geometric reference structure or at one or more angular offsets relative to the tangential geometric reference structure.

6. The system of claim 5, wherein the initial set of candidate radiation delivery geometries defines at least one of i) a set of radiation delivery beams and ii) a set of continuous radiation delivery arcs, as per the selected radiotherapy treatment modality.

7. The system of claim 1, wherein the subset of radiation delivery geometries selected for inclusion in the updated radiotherapy treatment pre-plan have minimized dose estimate value associated therewith.

8. The system of any one of claims 1 to 7, wherein the instructions stored on the memory further cause the system to perform operations that include:providing the updated radiotherapy treatment plan and the at least one treatment plan parameter to a treatment planning subsystem for performing forward dose calculation and dose normalization.

9. The system of claim 1 , wherein the surface facing portion of the target volume is proximate the patient surface; andwherein the at least one tangential geometric reference structure is defined through a point on the patient surface and has an orientation determined based on a local surface geometry of a part of the surface facing portion of the target volume that is closest to the surface of the patient.

10. The system of claim 4, wherein the pre-procedural image data includes a plurality of image slices of the patient that encompass the target volume.

11. The system of claim 10, wherein the at least one tangential geometric reference structure comprises a tangent line or tangent plane that is determined as an aggregate tangent for the target volume across a subset of the plurality of image slices which span a superior-inferior extent of the target volume.

12. The system of claim 10, wherein the at least one tangential geometric reference structure comprises a tangent line or tangent plane determined on one image slice that is defined through a superior-inferior midpoint of the target volume.

13. The system of claim 10, wherein the virtual, contoured anatomical structures define a three-dimensional representation of the target volume of the patient; and wherein defining the tangential geometric reference structure comprises reducing the three-dimensional representation of the target volume to a two-dimensional analysis on a representative image slice of the plurality of image slices that include the target volume.

14. The system of claim 13, wherein defining the tangential geometric reference structure further comprises:extracting a two-dimensional point set corresponding to a contour of the target volume on the representative image slice; andcomputing a tangent angle of the tangential geometric reference structure based on geometric extrema of the two-dimensional point set.

15. The system of claim 1, wherein the set of virtual, contoured anatomical structures further comprises at least one body contour defined in a portion of the nontarget volume that surrounds the target volume.

16. The system of claim 15, wherein the at least one treatment plan parameter comprises a plurality of treatment plan parameters; andwherein at least one of the plurality of treatment plan parameters includes a bolus parameters for a bolus dose of the radiation to be applied to the target volume of the patient.

17. The system of claim 16, wherein the instructions stored on the memory further cause the system to perform operations that include:defining a virtual bolus structure associated with the target volume based on the at least one bolus parameter;wherein the virtual bolus structure is utilized within the simplified dose calculation model to determine the respective dose estimate value for with each candidate radiation delivery geometry of the initial set of candidate radiation delivery geometries.

18. The system of claim 17, wherein defining the virtual bolus structure comprises expanding a virtual representation of the target volume outward toward the at least one body contour by an amount that is established from the at least one bolus parameter.

19. The system of claim 18, wherein the instructions stored on the memory further cause the system to: automatically generate one or more additional planning structures based on the target volume and the at least one body contour, wherein theone or more additional planning structures are used by the treatment planning subsystem during dose optimization and are not deliverable radiation structures.

20. The system of claim 19, wherein the one or more additional planning structures comprise a first tuning structure that surrounds the virtual bolus, the first tuning structure including a gap relative to the virtual bolus structure and extending to a predetermined depth on a deep side of the virtual bolus structure.

21. The system of claim 20, wherein the one or more additional planning structures comprise a second tuning structure that is offset from, and adjacent to, the first tuning structure, and extends deeper into the patient than the first tuning structure to limit intermediate-to-low irradiation outside the target volume.

22. The system of claim 21 , wherein the system is configured to constrain the dose optimization of the treatment planning subsystem using at least the first tuning structure and the second tuning structure as optimization constraint structures to thereby promote conformity of a high-dose region to the target volume and control dose falloff in the non-target volume of tissue.

23. The system of claim 21 or 22, wherein the at least one body contour includes a first body contour that immediately surrounds the target volume and a second body contour that surrounds the first body contour; andwherein the first body counter defines the first tuning structure, and the second body contour defines the second tuning structure.

24. The system of claim 6, wherein the radiotherapy treatment modality selected by the input of the user comprises intensity-modulated radiation therapy (IMRT);wherein the initial set of radiation delivery geometries comprises a set of fixed-angle radiation delivery beams; andwherein the at least one treatment plan parameter includes a desired number fixed-angle radiation delivery beams, a minimum separation angle between adjacent fixed-angle radiation delivery beams of the set of fixed-angle radiation delivery beams, and an angular offset relative to the tangential geometric reference structure.

25. The system of claim 24, wherein generating the initial set of radiation delivery geometries comprises placing the set of fixed-angle radiation delivery beams at angular positions defined relative to the tangential geometric reference structure.

26. The system of claim 25, wherein placing the set of fixed-angle radiation delivery beams at angular positions comprises:placing a first pair of fixed-angle radiation delivery beams at angular positions defined by the angular offset relative to the tangential geometric reference structure, wherein the first pair of fixed-angle radiation delivery beams are positioned on opposing sides of the tangential geometric reference structure; andplacing a remainder of the set of fixed-angle radiation delivery beams at angular positions that are spaced from one another and from the first pair of fixed-angle radiation delivery beams by at least the minimum separation angle.

27. The system of claim 6, wherein the radiotherapy treatment modality selected by the input of the user comprises volumetric modulated arc therapy (VMAT);wherein the initial set of radiation delivery geometries comprise the set of continuous radiation delivery arcs; andwherein the at least one treatment plan parameter includes a desired number of radiation delivery arcs in the set of continuous radiation delivery arcs and an angular offset relative to the tangential geometric reference structure.

28. The system of claim 27, wherein generating the initial set of radiation delivery geometries comprises defining, for each radiation delivery arc of the plurality of radiation delivery arcs, an arc start angle and an arc stop angle relative to the tangential geometric reference structure.

29. The system of claim 1 , wherein employing the simplified dose calculation model comprises:i) selecting one candidate radiation delivery geometry of the initial set of candidate radiation delivery geometries, the candidate radiation delivery geometry having an angular offset relative to the tangential geometric reference structure;ii) determining the respective dose estimate value for the candidate radiation delivery geometry;iii) removing the candidate radiation delivery geometry; andiv) iteratively repeating steps i) to iii) for each candidate radiation delivery geometry from the initial set of candidate radiation delivery geometries for generating the respective dose estimate values for all the candidate radiation delivery geometries;wherein the subset of candidate radiation delivery geometries are selected based on the respective dose estimate values.

30. The system of claim 29, wherein the candidate radiation delivery geometries comprise radiation delivery arcs, and wherein applying the radiation geometry optimization process comprises generating the radiation delivery arcs by varying at least one of an arc start angle or an arc stop angle by an angular increment across a predetermined search range.

31. The system of claim 29, wherein the subset of candidate radiation delivery geometries selected from the initial set delivery geometries are associated with lowest determined values of the respective dose estimate values.

32. The system of claim 29, wherein the subset of radiation delivery geometries are constrained to satisfy a predetermined, minimum separation angle between adjacent radiation delivery geometries of the set of radiation delivery geometries.

33. The system of claim 1, wherein the simplified dose calculation model derives the respective dose estimate value for each candidate radiation delivery geometry as a product of i) a peripheral path length through the non-target volume of tissue and ii) a projected area of the candidate radiation delivery geometry at the surface of the patient.

34. The system of claim 33, wherein the peripheral path length through the non-target volume of tissue is computed by:determining a beam path length through the patient between a beam entry point of the and a beam exit point to define a total path length PL;determining a path length through the target volume PLTarget along the beam path; anddetermining the peripheral path length as PLperiph = PL - PLTarget.

35. The system of claim 34, wherein the dose estimate value comprises PLperiph multiplied by the projected area.

36. The system of claim 1, wherein the instructions stored on the memory further cause the system to perform operations that include:generating machine-readable radiotherapy treatment plan data that configure a radiation delivery system to deliver radiation according to the selected subset of radiation delivery geometries.

37. The system of claim 1, wherein the instructions stored on the memory further cause the system to perform operations that include:applying the subset of radiation delivery geometries to constrain forward dose calculation and dose normalization in the treatment planning subsystem during the generation of the radiotherapy treatment plan data.

38. The system of claim 1 , further comprises a user interface that includes a display and that is operably connected to the processing hardware;wherein the instructions stored on the memory further cause the system to perform operations that include generating a visual rendering of at least part of the preprocedural image data and presenting the at least part of the pre-procedural image data to the user via the display.

39. A system for generating a radiotherapy treatment pre-plan for the delivery of a dose of radiation to a target volume of a patient, the system comprising:processing hardware comprising at least one processor and associated memory, the memory storing instructions that, when executed by the at least one processor, cause the system to perform operations that include:obtaining a set of virtual, contoured anatomical structures, at least some of the contoured anatomical structures defining the target volume of the patient, a nontarget volume of the patient, and a patient surface;receiving an input from the user that defines a radiotherapy treatment modality and at least one treatment plan parameter for the radiotherapy treatment pre-plan;defining, based on the target volume, at least one geometric reference structure associated with the target volume;employing the selected radiotherapy treatment modality, the at least one treatment plan parameter, and the at least one geometric reference structure to determine an initial set of candidate radiation delivery geometries defined relative to the geometric reference structure;employing a simplified dose calculation model to estimate, for each candidate radiation delivery geometry of the initial set of candidate radiation delivery geometries, a respective dose estimate value that corresponds to an irradiation of the non-target volume of the patient;selecting a subset of candidate radiation delivery geometries from the initial set of candidate radiation delivery geometries based on a reduction of dose delivered to the non-target volume; andproviding the selected subset of radiation delivery geometries to a treatment planning system that is configured to generate a radiotherapy treatment plan by performing a full dose calculation based on the selected subset of radiation delivery geometries;wherein the simplified dose calculation model that is less computationally complex than the full dose calculation performed by the treatment planning system and is executable using reduced computational resources and reduced execution time relative to the full dose calculation.

40. A system for generating a radiotherapy treatment pre-plan for the delivery of a dose of radiation to a target volume of a patient, the system comprising:processing hardware comprising at least one processor and associated memory, the memory storing instructions that, when executed by the at least one processor, cause the system to perform operations that include:obtaining a set of virtual, contoured anatomical structures, at least some of the contoured anatomical structures defining the target volume of the patient, a nontarget volume of the patient, and a patient surface;receiving an input from the user that defines a radiotherapy treatment modality and at least one treatment plan parameter for the radiotherapy treatment pre-plan;defining, based on the target volume, at least one geometric reference structure associated with the target volume;employing the selected radiotherapy treatment modality, the at least one treatment plan parameter, and the at least one geometric reference structure to determine an initial set of candidate radiation delivery geometries that are defined relative to the geometric reference structure;employing a simplified dose calculation model to estimate, for each candidate radiation delivery geometry of the initial set of candidate radiation delivery geometries, a respective dose estimate value that corresponds to an irradiation of the non-target volume of the patient;selecting a subset of candidate radiation delivery geometries from the initial set of candidate radiation delivery geometries for inclusion in an updated radiotherapy treatment pre-plan based on a reduction of dose delivered to the nontarget volume; andproviding the selected subset of radiation delivery geometries to a treatment planning system for the generation of a radiotherapy treatment plan;wherein the selecting of the subset of candidate radiation delivery geometries is performed without prompting the treatment planning system to perform iterative dose optimization.

Images