Virtual dose mapping for radiation processing

Abstract

A method for radiation processing of an object includes: accessing a virtual representation of the object including virtual elements; simulating radiation from a radiation source; perturbing the virtual representation to position and orient each of the virtual elements to generate a configuration of the virtual elements; and calculating a dose received by each virtual element in that different configuration. A method includes: generating virtual tiles that are located on a surface of the virtual representation; and simulating the radiation and calculating a dose received by each of the virtual tiles. A method includes: accessing a representation that includes a virtual object representing the object and a virtual surface that encompasses the virtual object; calculating a phase space distribution of radiation delivered to the virtual surface; and using the phase space distribution to calculate a dose received by the virtual object.

Description

VIRTUAL DOSE MAPPING FOR RADIATION PROCESSINGREFERENCE TO RELATED U.S. APPLICATION[00011 This application claims priority to the U.S. Provisional Application entitled “Adaptive Computer-Aided Design Models for Radiation Processing Simulations,” by T. Funk et al., Serial No. 63 / 580,560, filed September 5, 2023, hereby incorporated by reference in its entirety.BACKGROUND

[0002] Radiation processing involves exposing a product to an intense beam of radiation (for example, high energy photons or electrons) in order to influence the properties of the product. Radiation processing is widely used in a range of radiation processing applications, and one of the most prominent uses is the sterilization of medical devices and food.

[0003] One of the key parameters that needs to be specified in a radiation processing application is the amount of dose that will be delivered to the product. Dose is defined as the amount of energy deposited by the radiation per unit mass. In many radiation processing applications, the applied dose target is specified as a range, where all parts of the irradiated product are expected to receive a dose that is above a specified minimum dose and below a specified maximum dose. The reason for this is that some material properties can be modified by radiation exposure in undesirable ways (such as the yellowing of plastics), and so limiting the dose to a narrow range minimizes the chances of unwanted radiation damage. The dose range is often characterized by the Dose Uniformity Ratio (DUR), which is the ratio of the maximum and the minimum doses. Typically, a DUR close to 1.0 is preferred.

[0004] As an example, consider the development of a sterile medical device. Radiation sterilization is one of the “established” sterilization modalities defined by a regulatory guidance published by the United States Food and Drug Administration (U.S. FDA), which addresses gamma, electron beam (e-beam), and X-ray sterilization. Two of the important industry consensus standards for radiation sterilization are ISO 11737-1 and ISO 11137-2, which define the “sterilization dose” (i.e., the minimum dose necessary to achieve sterility) and the maximum acceptable dose that can received without compromising device functionality. Radiation sterilization is performed by irradiating a pallet or tote of fully packaged medical devices, which typically pass by the radiation beam on a conveyor belt. The dose received by every’ single device in the packed pallet must be within the dose rangedefined by the consensus standards; thus, the dose is affected by the packaging of each device, how the devices are arranged in their shipping boxes, and how the shipping boxes are arranged on the pallets.

[0005] Ensuring that the dose delivered to all products is within the specified dose range is ty pically part of the qualification process of a radiation processing application, and is even a regulatory requirement for medical device sterilization. To do this, manufacturers work with radiation processing facilities to establish the delivered maximum and minimum doses by performing a series of measurements on individual products and products packaged in shipping boxes. Dosimeters are placed at various positions on the products, with educated guesses by the radiation processing technicians as to where the minimum and maximum dose are expected to occur. Measurements are then repeated several times: first at different orientations of the product relative to the beam, and then several times at the chosen orientation to establish the measurement uncertainty. This process can take several weeks to complete and requires measurements of dozens of manufactured products. This can be very' expensive at the point of process validation, as the manufacturing process is likely to have not been fully established.

[0006] Dosimeters are often small pieces of radiochromic film. These dosimeters have a substantial area (for example, one [1] centimeter [cm] by 1 cm), and so they measure the average dose over their extent. The substantial size of the dosimeters limits their ability to measure the dose in small structures, such as the interiors of needles. There is currently no way to measure the dose in such restricted geometries. Furthermore, dosimeters only measure the dose at the location where they are placed; that is, although they allow for spot checks of the dose distribution, they cannot measure the full, three-dimensional dose distribution. An additional limitation of dosimeter measurements is that they can only measure the dose over a limited range of dose. For example, the response of many radiochromic films degrades below five kilogray' (kGy).

[0007] A further challenge of using dosimeters to measure the dose distribution is the uncertainty in the positioning of the dosimeters. There are two factors that give rise to this uncertainty. First, dosimeter measurements are typically repeated multiple times in order to sample the repeatability of the measurement. Attempts are made to place the dosimeters in the same positions on the product for each measurement, but there are always positioning errors that give rise to uncertainty in the measured dose. The second cause of the uncertainty is that the positions of the dosimeters on the product are often poorly documented. It is typical that a photograph of the dosimeter affixed to the product is not taken, and instead onlya marker on a photograph of the product is used to indicate roughly where the dosimeter was placed.SUMMARY

[0008] A family of virtual representations of an object, configured to mimic how elements of the object are typically arranged for radiation processing, is generated in order to capture the variations between the object and its virtual representation. Some of the surfaces of the virtual object are tiled with virtual dosimeters to allow for the calculation of a spatial surface dose distribution.

[0009] In embodiments according to the present invention, a computer-implemented method for radiation processing of an object includes: accessing a virtual representation of the object, where the virtual representation includes a plurality of virtual elements; simulating radiation from a radiation source; perturbing the virtual representation to position and orient each of the virtual elements to generate a configuration of the virtual elements; and calculating a dose received by each virtual element in that different configuration.

[0010] In other embodiments, a computer-implemented method includes: generating virtual tiles that are located on a surface of the virtual representation of the object, where the virtual tiles are at locations on the surface corresponding to locations on the object that are in contact with an environment surrounding the object; and simulating radiation from a radiation source and calculating a dose received by each of the virtual tiles.

[0011] In yet other embodiments, a computer-implemented method includes: accessing a representation that includes a virtual object representing the object and a virtual surface that encompasses the virtual object; calculating a phase space distribution of radiation delivered to the virtual surface; storing the phase space distribution in computer system memory; and simulating radiation from a radiation source using the phase space distribution to calculate a dose received by the virtual obj ect.

[0012] The features of the embodiments presented above, along with features described in the detailed description to follow, can be utilized in other embodiments that include those features in any practical or possible combination.

[0013] Embodiments according to the present invention overcome the shortcomings of conventional radiation processing applications that are described above. Accordingly, radiation processing can be performed less expensively and more quickly, and without the necessity of physical dosimeters and the burden of their associated limitations and challenges.BRIEF DESCRIPTION OF DRAWINGS

[0014] The present invention is illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which like reference numerals refer to similar elements.

[0015] Fig. 1 is a flow diagram illustrating a computer-implemented method of generating a family of virtual dose maps corresponding to a group of virtual representations of products in embodiments according to the present invention.

[0016] Fig. 2 is a diagram showing an example of a virtual configuration of packaged products packed for radiation processing in embodiments according to the present invention.

[0017] Fig. 3 is a diagram showing an example of a product with a flexible component that can have multiple configurations in embodiments according to the present invention.

[0018] Fig. 4 is a diagram illustrating an example of how simulated shaking a virtual representation of a packaged product can produce a new configuration, in embodiments according to the present invention.

[0019] Fig. 5 is a flow diagram illustrating a computer-implemented method of generating the surface dose distribution by meshing the surface and interior of a virtual presentation of a product with tiles in embodiments according to the present invention.

[0020] Fig. 6 is a diagram illustrating examples of virtual meshing of a surface and interior of a virtual representation of a product, and a demonstration of calculating the dose delivered to such a virtual meshing, in embodiments according to the present invention.

[0021] Fig. 7 is an illustrative example of a radiation processing facility showing the path taken by totes carrying a product to be irradiated, in embodiments according to the present invention.

[0022] Fig. 8 is a flow diagram illustrating a computer-implemented method of generating a phase space distribution to model an irradiation source for subsequent virtual dose mapping simulations, in embodiments according to the present invention.

[0023] Fig. 9 is a diagram illustrating an inverse method of calculating the phase space distribution on a virtual surface around a tote, in embodiments according to the present invention

[0024] Fig. 10 is a block diagram illustrating an example of a computer system on which embodiments according to the present invention can be implemented.DETAILED DESCRIPTION

[0025] Embodiments according to the present invention pertain to methods to prepare virtual (digital) models to be used for simulating the dose delivered in radiation processing applications.

[0026] Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings. While the invention will be described in conjunction with these embodiments, it will be understood that they are not intended to limit the invention to these embodiments. On the contrary, the invention is intended to cover alternatives, modifications, and equivalents, which may be included within the spirit and scope of the invention as defined by the appended claims. Furthermore, in the following detailed description of embodiments of the present invention, numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, it will be recognized by one of ordinary skill in the art that the present invention may be practiced without these specific details. In other instances, well-known methods, procedures, components, and circuits have not been described in detail as not to unnecessarily obscure aspects of the embodiments of the present invention.

[0027] It will also be understood that the present invention will be described in terms of a given illustrative architecture; however, other architectures, structures, materials, and process features and steps may be varied within the scope of the present invention.

[0028] Some portions of the detailed descriptions that follow are presented in terms of procedures, logic blocks, processing, and other symbolic representations of operations on data bits within a computer memory7. These descriptions and representations are the means used by those skilled in the data processing arts to most effectively convey the substance of their work to others skilled in the art. In the present application, a procedure, logic block, process, or the like, is conceived to be a self-consistent sequence of steps or instructions leading to a desired result. The steps are those utilizing physical manipulations of physical quantities. Usually, although not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated in a computer system. It has proven convenient at times, principally for reasons of common usage, to refer to these signals as transactions, bits, values, elements, symbols, characters, samples, pixels, or the like.

[0029] It should be bome in mind, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless specifically stated otherwise as apparent from thefollowing discussions, it is appreciated that throughout the present disclosure, discussions utilizing terms such as “generating,” “storing,” “calculating,” “positioning,” “displaying.” “orienting,” “combining,” “saving,” “arranging,” “creating,” “perturbing,” “deriving,” “varying,” “grouping,” “identifying,” “producing,” “determining,” “accessing,” “representing,” “simulating,” “modifying,” “using,” or the like, refer to actions and processes of an apparatus or computer system or similar electronic computing device or system (e.g.. the computer of Fig. 10). A computer system or similar electronic computing device manipulates and transforms data represented as physical (electronic) quantities within memories, registers or other such information storage, transmission or display devices.

[0030] Some elements or embodiments described herein may be discussed in the general context of computer-executable instructions residing on some form of computer- readable storage medium, such as program modules, executed by one or more computers or other devices. By way of example, and not limitation, computer-readable storage media may comprise non-transitory computer storage media and communication media. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform particular tasks or implement particular abstract datatypes. The functionality of the program modules may be combined or distributed as desired in various embodiments.

[0031] Computer storage media includes volatile and nonvolatile, removable and nonremovable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to, random access memory (RAM) or dynamic RAMs (DRAMs), read only memory (ROM), electrically erasable programmable ROM (EEPROM), flash memory (e.g., an SSD) or other memory technology, compact disk ROM (CD-ROM), digital versatile disks (DVDs) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and that can accessed to retrieve that information.

[0032] Radiation transport simulations take a digital model (e.g., a Computer-Aided Design [CAD] model) of a product being irradiated as an input, and output a calculated, three-dimensional dose distribution. Simulations address many of the limitations of dosimeter measurements, because simulations can calculate the full three-dimensional distribution of dose and can have extremely fine spatial resolution. Generally speaking, as used herein, a “product” is any object that can and may be subject and subjected to radiation processing, including devices, and which may comprise a single component or part or multiplecomponents or parts. An object may be physical object that exists in the real world, or it maybe a virtual object that has not been physically manufactured. That is. in the latter case, the object may exist only in the abstract or virtual world (e.g., in the CAD model, as an embodiment at one stage of a design process of a “work in progress,” or as an embodiment at the final stage of the design process, prior to manufacture). The term “object” may be used herein as being equivalent to the term “product.”

[0033] Representing dosimeters directly in CAD models for radiation simulations is a possibility; for instance, dosimeters can be modelled as, for example, two-dimensional rectangles or three-dimensional rectangular prisms with the same sizes as the real radiochromic film (e g., 1 cm x 1 cm x 0.18 millimeters [mm]). Similarly, alanine pellet dosimeters can be modelled as small cylinders. Conventional techniques for creating such virtual dosimeters are manual processes that can be very time-consuming; for example, if the surface of the product to which the virtual dosimeter is to be affixed is curved, then the dosimeter must be modelled with the same curvature in order to accurately model how a real dosimeter would be affixed to the product. Radiation simulations with virtual dosimeters modelled in this way can calculate the total dose that would be delivered to a dosimeter in that position. However, virtual dosimeters modelled in this way- have a number of limitations: first of all, the calculated dose is based on an assumption that the real dosimeter was exactly- placed as it was in the CAD model. However, due to gradients in the three-dimensional (3D) dose distribution, even small variations in the dosimeter placement can result in large changes in the calculated dose. Because placing physical dosimeters on products is a manual process, there will always be some discrepancy between where the virtual dosimeter is modelled and where the real dosimeter w as placed.

[0034] For conventional radiation processing applications, the physical products that are exposed to radiation can differ significantly from the corresponding CAD models that were developed in the design phase of the product development process. There are several causes of this. For instance, many packaging designs do not constrain the product to a single position and orientation. As an example, many medical devices are packaged in plastic pouches, which allow for significant variation in the positioning and orientation of the device within the pouch. An additional cause for the discrepancy^ between the CAD model of a physical product and the physical product itself is that many products contain flexible components, such as tubing, that can have significant variations in their arrangements from product-to-product. The consequence of these factors is that, even for identical products, each physical and packaged version of those products may have a slightly different configurationwhile undergoing radiation processing, and each of which will differ to some degree from the idealized CAD model of the product.

[0035] Because conventional radiation simulation tools typically take a CAD model as their input, the calculated dose distribution will correspond precisely to the configuration of the CAD model that is input. Because each of the real, physical products can differ from its CAD model due to the factors discussed above, the dose distributions calculated by radiation simulations will correspondingly differ from the “true7’ dose delivered to each physical product, because the dose distributions depend on the specific configuration of the physical product being irradiated. Therefore, the real, physical variations in the product / packaging configurations limits the usefulness and predictive ability of such conventional radiation modelling tools for radiation processing applications.

[0036] With respect to radiation sterilization, these limitations pose a significant risk when attempting to use radiation simulations as a predictive tool to inform the sterilization process. This is due to several reasons. First, because simulations are typically only performed on a single CAD model that is supposed to represent each of a number of identical physical products, only a single dose distribution is calculated, and it is assumed that this dose distribution is what is delivered to each of those products. However, due to the physical variations between the individual, identical products, it is possible that the dose uniformity' ratio (DUR) and the location and magnitude of the minimum and maximum doses vary from product-to-product. These variations would be completely’ missed using conventional radiation simulations as a predictive tool. These variations pose a risk to the sterilization validation process, and ultimately to the health of patients, because some of the products might fall outside the acceptable dose range. Second, the variations in the delivered dose distributions may indicate that the terminal sterilization is suboptimal from the process perspective. For instance, a different packaging design might reduce the dose variations, thus leading to a higher quality and efficiency in the sterilization process.

[0037] Embodiments according to the present invention provide solutions to the above problems. In these embodiments, computational methods are used to automatically produce a family of related digital (virtual) representations of products that are to be used for radiation processing simulations. These virtual models contain model components that correspond to how the products are typically configured for radiation processing; for example, a typical configuration involves placing packaged products in shipping boxes, and arranging the shipping boxes on pallets. Conventional radiation simulations suffer from the shortcoming that the virtual models they are based on can differ significantly from the actualphysical realization and locations / orientations of the products they are supposed to represent. This can severely hinder the predictive ability of conventional radiation simulations. Embodiments according to the present invention overcome this limitation by including steps to vary the configuration of the virtual models and generate a family of related virtual models that represent the variations that are experienced by the physical realization of the products. This way, rather than solely considering only the dose distribution of one idealized configuration, embodiments according to the present invention consider the dose distributions in multiple (and perhaps all) meaningful configurations of the product.

[0038] A method in an embodiment according to the present invention is illustrated in Fig. 1. In step 100, a representation of an object is accessed from computer system memory. In an embodiment, a CAD model of the object is loaded. The CAD model is a three- dimensional, virtual representation of an object or element (e.g., a physical object or an element / part of a physical object, or a virtual object or an element / part of the virtual object). In embodiments according to the present invention, a virtual representation that mimics a configuration that is commonly used for radiation processing of the object is used.

[0039] Fig. 2 is a diagram showing an example of such a configuration. In the example of Fig. 2, a tote or pallet 200 is packed with an arrangement of shipping boxes 201, and each shipping box contains a number of products 202 in their packaging 203. As mentioned above, the products 202, packaging 203, and shipping boxes 201 may be physical objects that exist in the real world, or they may be virtual objects. In other embodiments according to the invention, the virtual representation contains, for example, a single product 202 in its packaging 203, or a set of packaged products 202 / 203 in a single shipping box 201, or a set of shipping boxes 201 without a tote or pallet 200, or combinations thereof. Other embodiments according to the invention use a virtual representation that consists of, or includes, a set of packaged products 202 / 203 directly on the pallet / tote 200. Other virtual representations may include multiple nested layers of packaging 203.

[0040] In embodiments, a single CAD model is used to virtually represent the elements in the entire packing configuration in the example of Fig. 2 (e g., a single CAD model that contains all packed shipping boxes 201 on a pallet 200). However, other embodiments according to the invention use multiple CAD models to represent the configuration that will be exposed to radiation. For example, in such other embodiments, each element in the configuration (e.g., the product 202, its packaging 203, the shipping box 201, and the pallet 200) can be represented by a separate, respective CAD model. In such cases, the method illustrated in Fig. 1 includes a step 101 that combines all of the individualCAD models into a single CAD model. In some embodiments, the step 101 is performed manually by a user (e.g.. an engineer or designer) using CAD software. However, in other embodiments according to the invention, the step 101 is performed automatically using an algorithm that integrates the individual CAD models into a single CAD model. Thus, generally speaking as used herein, a CAD model may be a single, monolithic model representing the whole of a number of different elements (e g., the product 202, its packaging 203, the shipping box 201. and the pallet 200); or the monolithic CAD model may be a single model that integrates a number of component CAD models that each represents a subset of one or more of those elements.

[0041] For example, consider the case when the product 202 and its packaging 203 are represented with separate, respective CAD models. This case can arise for products developed by large companies, where different departments are responsible for the design of the product and the packaging. In this case, the disclosed method allows the user to identify which of the CAD models corresponds to the packaging 203. Then, a void (e g., an empty volume) within the CAD model for the packaging 203 can be automatically identified. If the void is large enough to encompass (fully contain) the CAD model (virtual representation) of the product 202, then the product 202 can be automatically placed within the void. If multiple orientations of the product 202 can fit within the void (e.g., if the void allows for rotations of the product 202, or for the product 202 to be flipped around one of its axes, etc.), then multiple configurations of the combined CAD model of the product 202 and its packaging 203 are generated in step 101 . If the void is not large enough to encompass the product 202, then the user is notified and prompted to modify the CAD model of the packaging 203 or the CAD model of the product 202.

[0042] If multiple CAD models are used to represent the various elements of the virtual representation shown in Fig. 2, then, in some situations, the elements are not arranged in a repeating pattern, as they are in the example of Fig. 2 (that is, the packaged products 202 / 203 may be oriented differently in different shipping boxes 201, and / or the shipping boxes may be arranged differently on the pallet 200, versus what is show in the figure). In this case, an algorithm may be used to automatically pattern the virtual elements representing the packaged products 202 / 203 in the shipping boxes 201, and / or to automatically pattern the virtual elements representing the shipping boxes 201 on the pallet 200. In general, in an embodiment, the virtual elements are automatically rearranged (spatially) according to a prescribed pattern. In an embodiment, an optimization algorithm is applied (with specified constraints) to vary the positions and orientations of the virtual elements (e.g., the virtualelements representing the packaged products and the shipping box) until an optimization problem is satisfied. For example, a volume (e.g., a box-shaped volume) that bounds each packaged product 202 / 203 can be used to calculate how many packaged products can fit in a single shipping box 201 by, for example, considering 90-degree rotations of that bounding box along each of the three principal axes. This methodology can similarly be used to automatically pattern shipping boxes 201 on the pallet 200.

[0043] In other embodiments, the user does not have a CAD model for one of the elements that they wish to model. For example, they may have a CAD model of the product 202 but not a CAD model of the packaging 203. In this example, a CAD model of the packaging 203 can be automatically created based on the shape of the product 202. This may involve using constructive solid geometry (CSG) to model the packaging 203 using Boolean combinations of simple shapes, also known as geometric primitives (e.g., boxes, spheres, cylinders, prisms, pyramids, cones, etc.). Alternatively, a more sophisticated approach can be used, in which an optimization algorithm, such as a genetic algorithm, is used to automatically design the shape of the packaging 203 given the constraint of the shape of the product 202, as well as any external constraints that are chosen by the user (including, but not limited to, specifying the minimum allowed space betw een the product 202 and the packaging 203, limiting the maximum size of the void in the packaging 203, etc.). This approach can be similarly applied to any component that the user wishes to model but does not have a CAD model for (e.g.. the shipping box. pallet, etc.). Generally speaking, in an embodiment, a model (virtual representation) of a first virtual element can be created based on a specified constraint; for example, the specified constraint is that the first virtual element has a void (volume) large enough to fully contain a second virtual element.

[0044] Once a single CAD model of the configuration that will be exposed to radiation has been created as in step 101, the elements represented by the model (virtual elements) are perturbed in a step 102 to generate a number of unique and physically possible variations of the configuration, by vary ing the position and orientation of some or all of the elements represented by the CAD model. The representation is perturbed to position and orient each element in the CAD model, to generate a configuration of those elements. By perturbing the virtual elements, the virtual elements can be rearranged, repositioned, and / or reoriented relative to their initial configuration and relative to subsequent configurations resulting from the perturbing. In an embodiment, geometric transformations that achieve a resulting position and orientation of each virtual element in each of the different configurations are saved as entries in a table in computer system memory (Fig. 10). In anembodiment, the virtual elements are automatically rearranged (spatially) according to a specified pattern prescribed in advance of the perturbing step.

[0045] The perturbations take into account the physical constraints of the packing configuration. For example, the product might be a single object within its packaging. If the size of the packaging is large enough relative to the product, then the product might be essentially unconstrained and free to take on any orientation or position within the bounds of the packaging. Alternatively, the packaging may partially constrain the product, by restricting the rotational degrees of freedom of the product, the translational degrees of freedom of the product, or both the rotational and translational degrees of freedom. Finally, the packaging may fully constrain the product when no movement of the product within the packaging is possible.

[0046] Some products might consist of multiple components. This is illustrated in Fig. 3, which shows an example of a product that includes three components: a flexible tube 300 and two end caps 301. For each component of a product, the packaging may partially or fully constrain that component relative to one or more of the other components. Furthermore, the packaging may apply different levels of constraint to different components. For example, one component of the product may be fully constrained so that it does not have any rotational or translational degrees of freedom within the packaging, while another component of the product may be able only partially constrained, allowing it to move around within the packaging relative to the fully constrained component of the product.

[0047] For products with multiple components, the components may be connected or unconnected. For instance, consider a pinwheel toy packaged in a cardboard box. The pinwheel is composed of two connected components: the handle, and the wheel. The handle of the pinwheel may be constrained by fasteners that restricts it to a single position within the box. However, the pinwheel may be unconstrained, and thus allowed to freely rotate within the packaging.

[0048] Additionally, as discussed above, a ty pical configuration for radiation processing includes multiple packaged products in shipping boxes, and multiple shipping boxes placed on a pallet. Each of these elements can move relative to each other; for example, each individual shipping box may move relative to each other and to the pallet, and each packaged product within a shipping box may move relative to the other packaged products in the shipping box.

[0049] If multiple elements (e.g., products, or components / parts / elements of products) are within the same packaging, they will also place constraints on each other, as they cannotoccupy the same space. For example, some of the elements may abut (be in contact with) one or more other elements. They may also interact with each other through collisions. These collisions may be modelled with rigid-body dynamics, in which case the shape of the elements represented by the CAD models do not change. Alternatively, the collisions may be modelled with soft-body dynamics, in which case the shape of the modelled elements can change. The latter case is particularly pertinent when some of the components of the CAD model represent elements that are flexible or deformable. For example, some packaging 203 may be fully flexible, such as a pouch, or have a part of the packaging that is flexible. Similarly, part of the product 202 itself may be flexible. This is show n in the example of Fig. 3. The end caps 301 may be modelled as rigid bodies, whereas the flexible tube 300 may be modelled as a soft body. Due to the deformable nature of the tube 300, multiple configurations 302 of the tube may be possible, assuming that they are allowed by any constraints due to the packaging (if it is present). In general, the different components of the CAD model can be represented as either rigid bodies or soft bodies.

[0050] In the case of either rigid-body or soft-body collisions, the interactions of the various elements represented in the CAD model depend on the material properties as well as the geometry of the element(s) represented by the model.

[0051] Another important aspect of perturbing the CAD model in step 102 of Fig. 1 is the external forces that are present. For example, gravity constrains where the product 202 can be in the packaging 203. Elastic properties of the product 202 or the packaging 203 also can exert forces on the product 202 and affect possible positions of the product 202. In some embodiments according to the invention, external forces are applied to the various components of the CAD model.

[0052] All the variations described above are examples of some of the many possible ways that the product and its components can move within its packaging and can be included in the perturbations of the CAD model applied in step 102 of Fig. 1. In the following paragraphs, various embodiments according to the invention, with different ways to achieve the required perturbations, are discussed.

[0053] In an embodiment, several possible configurations are generated by an engineer / designer using CAD software such as but not limited to SolidWorks® or Creo®. In this embodiment, the CAD model is loaded into the CAD software, and the CAD model of the product 202 is placed in an orientation and position within the constraints of the packaging 203. Depending on whether or not the CAD model also includes the shipping boxes 201 or pallet 200 as previously described herein, the CAD model of the productpackaging 203 may also be positioned relative to the shipping boxes 201, and the shipping boxes 201 positioned relative to the pallet / tote 200. This positioning may be done manually in the CAD software, or using a modelling tool included in the CAD software. The updated CAD model, with the new positions and orientations, is exported in a format compatible with radiation simulation software in step 103 of Fig. 1. This process is repeated until two, three, four, or any other number of physically possible configurations of the virtual elements presented by the CAD model have been generated.

[0054] In another embodiment of this invention, a physical realization of the product 202 and the packaging 203 is used to realize different configurations in step 102. To achieve this, the product 202 and packaging 203 are physically handled as they would be during packaging or shipping to a radiation processing facility, for example, as described in the consensus standard ASTM D4169. For example, this may include shaking the packaged product 202 / 203, actually stacking packed shipping boxes 201 onto a pallet 200, or transferring the packed pallet 200 down a conveyor belt. Once completed, an imaging modality, such as but not limited to computed tomography (CT), laser scanners, or magnetic resonance imaging (MRI), can be used to scan the physical realization of the packaged product 202 / 203. The output of the scan (e.g., a voxelized volume) can be used to define a virtual representation of the package product 202 / 203, which then be used either directly in radiation simulations or segmented into the different elements of the packaged product configuration, in which case the elements may be converted into surface or volume meshes. Different configurations of elements can be generated by repeating the above steps.

[0055] Because experimental dose mapping aims to measure the minimum and maximum doses in order to ensure that they are within the acceptable dose range, in another embodiment, step 102 includes using a computational optimization tool to identify the ■‘worst-case” configurations from a dose map perspective. This is similar to a tolerance stack- up analysis that is used in engineering. The tool can use a step-based optimization algorithm that tries to maximize the DUR (which would be the worst-case), where a new virtual configuration is generated at each step by changing the positions and orientations of some of the virtual elements of the configuration, and radiation simulations are used to calculate the dose distribution corresponding to the configuration at that step.

[0056] Because radiation simulations are relatively time consuming (typically taking minutes or hours), instead of using the calculated dose distribution as part of the optimization algorithm, a proxy metric that is much faster to calculate may be used. For example, because the dose distribution is roughly proportional to the mass density’ distribution, the optimizationcan maximize the inhomogeneity in the mass density distribution. Thus, in an embodiment, a mass density distribution based on the configuration is determined, and the dose distribution is modelled as a function dependent on that mass density distribution. This is much faster than calculating the DUR, because a radiation simulation does not need to be performed; instead, after the candidate virtual configuration is generated by modifying the positions / orientations of some or all of the virtual elements represented by the CAD model, then only the CAD model needs to be analyzed to quantify the inhomogeneity in the density.

[0057] In another embodiment, the different configurations of the virtual representation of the packed product will be realized using physics-based modelling software such as Blender®. Different virtual configurations can be realized by simulating shaking, using the CAD model, in consideration of forces such as gravity or from elastic deformations acting on the product. This simulated shaking is intended to model the jostling of the packaged physical product that would be typical during radiation processing including shipping and handling of the packages.

[0058] An illustration of an example of the process used to ’shake" the CAD model (that is, to simulate shaking of an object modelled by the CAD model) is illustrated in Fig. 4, which shows a product 400 in its packaging 401, in which the product can move unconstrained. Shaking can be modelled in physics-based modelling software by applying external forces 402 with random directions to the CAD model. In embodiments, the external forces 402 are derived from hazard elements, such as drops and vibrations, that are expected to occur during handling and transportation of the product 400 to, within, and from a radiation processing facility. Gravity’ 403 may also be modelled as an external force. In some embodiments according to the invention, the external forces 402 are sampled from the vibrational distributions defined in the consensus standard ASTM D4169, which describes the transportation of shipping containers and systems. ASTM D4169 also describes drop tests of shipping containers, which are simulated by using gravity 403 as an external force 402 as part of step 102 in Fig. 1 in some embodiments. Once the external forces are stopped and the velocity of the components of the CAD model of the product 400 are zero, then the final positions and orientations of the elements represented by the CAD model are taken as the final configuration 404. During the simulated shaking, the collision detection capabilities of the physics-based modelling software are used to identity’ collisions 405, which ensure that the physical product 400 remain within the packaging 401 without interferences. Because the simulated shaking is randomly applied, each virtual configuration will be uniquely generated, and so a family of unique virtual configurations can be generated this way. In anembodiment, statistical considerations are applied to eliminate certain virtual configurations from being included in the family. For example, if the probability of a configuration of virtual elements does not satisfy a prescribed probability threshold (e.g., it is unlikely to occur according to a prescribed statistical metric), then that configuration is not included in the family.

[0059] After a family of unique virtual configurations is generated in step 102 of Fig. 1 using any of the methods described above, some embodiments according to the invention perform radiation simulations to calculate the three-dimensional dose distribution for each configuration. In embodiments, radiation from a radiation source is modelled (simulated), and the dose produced by that radiation is calculated, to determine a three-dimensional dose distribution at (across) each of the elements (e.g., the product 202, its packaging 203, the shipping box 201, and the pallet 200 of Fig.2) represented in the CAD model. This process is known as virtual dose mapping and is shown as step 104 in Fig. 1, and can be performed on each of the families of virtual configurations. A statistical analysis of the resulting calculated three-dimensional dose distributions, also known as virtual dose maps, can be performed (step 105 in Fig. 1). The statistical analysis may involve the removal of statistical outliers, calculating the value and location of the minimum and maximum doses in the dose distribution, calculating a histogram of the dose, and calculating the DUR, for example. The statistical analysis can separately consider the dose delivered to the surfaces of all elements (e.g.. products), which is of interest for radiation sterilization, and the dose delivered to the bulk materials within the elements, which is important for general radiation processing. Furthermore, the analysis can segregate the data by different levels in the assembly hierarchy in the input CAD model; for example, the data can be segregated by product (e.g., to consider the distribution of dose at the product level) or can be segregated by shipping box (e.g., to consider the distribution of dose at the shipping box level).

[0060] In step 106, a similar analysis can be performed on the family of virtual dose maps, each of which corresponds to a unique configuration of the virtual representation of the product(s) under exposure to radiation. That is, the absolute minimum and maximum doses over all virtual dose maps can be calculated, along with the corresponding DUR. This global analysis can similarly be segregated at the product or shipping box level, and can also consider either the surface dose or the bulk dose.

[0061] In addition to creating a family of virtual configurations that facilitate realistic simulations of radiation processing applications, embodiments according to the invention also include a method to create virtual dosimeters to facilitate the use of radiation simulations toguide / analyze experimental dose mapping. Distinct virtual dosimeters are conventionally limited to calculating the dose at the precise location where they are included in the radiation processing model and methodology disclosed herein. They are also faced with a number of other challenges, such as the fact that it is difficult to model them.

[0062] Embodiments according to the invention overcome these limitations byproviding a method to “tile” some or all of the surfaces of the virtual product with small dosimeter tiles. By calculating the dose delivered to each tile using radiation simulations, embodiments according to the present invention provide a way to calculate a spatial surface dose distribution. The shapes and sizes of the tiles determine the spatial resolution of the simulated surface dose distribution, and different embodiments according to the invention use different methods to determine the tile shapes and sizes.

[0063] A method is illustrated in Fig. 5. a representation of the object / product (a virtual object) is accessed from computer system memory as in step 100 of Fig. 1. Because real dosimeters are affixed to the surfaces of the products being experimentally dose mapped, in step 500, the “unmated” surfaces of the virtual representation illustrated in Fig. 2 are identified. In CAD models, “mated” surfaces are simply two surfaces that are in contact; thus, the “unmated” surfaces are the surfaces that are in contact with the surrounding environment (e.g., air), and correspond to the potential surfaces that dosimeters can be attached to. These surfaces may be on the interior or the exterior of the virtual representation of the product. In step 501. virtual tiles are located on a surface of the representation of the object; the virtual tiles are at locations on the surface corresponding to locations on the object that are in contact with an environment surrounding the object (the unmated surfaces). In step 501, in an embodiment, some or all of the unmated surfaces are then covered with virtual 3D tiles. These virtual tiles can be included in the CAD model. In step 502, radiation from a radiation source is simulated, and a dose received by each virtual tile is calculated. By calculating the dose received by each tile using radiation simulations as in step 502, the spatial distribution of the doses on the tiles can be calculated.

[0064] The process described by Fig. 5 is illustrated in Fig. 6. where a CAD model 600 has its unmated surfaces covered in virtual surface tiles 601.

[0065] In the radiation simulations, the virtual tiles 601 can be assigned materials corresponding to typical dosimeter materials (e.g., radiochromic film). Alternatively, because most dosimeters are calibrated to the ratio of dose-to-water, the tiles 601 can be assigned the material of water. In other embodiments according to the invention, the virtual tiles are assigned the material of the environment surrounding the object, such as air. In general, eachvirtual tile 601 can be assigned a different material or a different density (e g., increased density to increase the radiation absorption cross section). The materials may or may not correspond to real-world materials; for example, for the simulation, the material may have its density artificially increased to increase the radiation absorption cross-section as a way to force more interactions between the incident radiation and the material, thus improving the statistical uncertainty of the calculated dose delivered to that material.

[0066] In some embodiments, the tiling uses the mathematical technique of “tessellation,” or meshing, where a surface is covered with virtual tiles 601 that have no overlaps and no gaps between them. In other embodiments, the virtual tiles 601 may overlap or may have gaps between them. As long as the overlapping virtual tiles 601 are assigned the same material in radiation simulations, overlapping tiles will not result in simulation artifacts.

[0067] Some tessellations may be “regular,” meaning all the tiles 601 have the same shape, whereas others may be “semiregular,” meaning that the tiles 601 are selected from a set of different shapes. Other embodiments use a different shape for each tile 601.

[0068] In some embodiments, the virtual tiles 601 have primitive shapes such as rectangular prisms or tetrahedra.

[0069] In some embodiments, the virtual tiling is calculated in an adaptive way, where the tile size is determined based on the size of the features on the corresponding virtualized product surface, such that smaller features are tiled with smaller tiles.

[0070] Some tiling may be periodic with a repeating pattern, whereas others may be aperiodic, which includes tile shapes that cannot form a repeating pattern.

[0071] In some embodiments, the thickness of the virtual surface tiles 601 is configured based on an input from a user. In other embodiments, the thickness of the surface tiles 601 varies from part to part within the product.

[0072] In some embodiments, the virtual surface tiling is thickened, either by using thicker virtual tiles 601 or by having multiple layers of the virtual tiles 601, such that the surface tiling extends into the volume (e.g., air) surrounding the product.

[0073] Although virtual tiles 601 are useful to calculate the spatial distribution of dose that is sampled during experimental dose mapping measurements, other embodiments according to the invention also tile the interior of some or all of the parts of the product into a 3D volumetric mesh of virtual interior tiles 602. In such embodiments, the interior tiles 602 are tetrahedra or other polyhedral shapes, although the invention is not so limited. Similar to the virtual surface tiles 601, the size and shape of the virtual interior tiles 602 can be calculated in an adaptive way depending on the size of the feature they are meshing.

[0074] Once the doses to the tiles have been calculated using radiation simulations (e.g., the dose distribution 603), embodiments according to the invention enable different analyses to be performed, as shown in step 504 of Fig. 5. For example, a histogram of tile doses can be calculated, along with statistics such as the minimum tile dose, maximum tile dose, and DUR. Additionally, the tile locations of the minimum and maximum tile doses can be identified.

[0075] In other embodiments, the analysis involves averaging the calculated dose delivered to a group of neighboring virtual tiles 601 (Fig. 6). The group of tiles can be used to represent a virtual dosimeter at the location of the group. The location and the size of the group of virtual tiles 601 can be varied to understand their impact on the calculated dosimeter dose. The average calculated dose can be calculated by a spatial filter (e.g.. a smoothing filter) that accounts for the curvature of the surface of the product / object. In other embodiments, various other types of spatial / smoothing filters are applied to the neighborhood of virtual surface tiles 601 prior to the analysis of step 504 of Fig. 5 (e.g., in step 503), such as but not limited to a median filter, an average filter, a Gaussian filter, or a bilateral filter. Calculating which tiles belong in the neighborhood of the spatial filter can involve calculating the distance between two surface tiles 601 using graph theoretical algorithms such as Dijkstra's algorithm or breadth-first search. To facilitate the use of such algorithms, the neighbor information between the virtual surface tiles 601 may be stored in a data structure such as an adjacency list or an adjacency matrix.

[0076] Additionally, the spatial dose gradient of the tile doses can be calculated. Because the virtual surface tiles 601 can represent surfaces with curvature, the dose gradient can be calculated as a directional derivative. The dose gradient can also be analyzed in multiple ways; in some embodiments, the dose gradient is used to refine the size of the surface tiling, so that a new tiling is generated with smaller virtual tiles 601 in regions with a higher dose gradient magnitude, and larger virtual tiles 601 in regions with a lower dose gradient. In other embodiments, the dose gradients are used to identify regions with low7gradient magnitudes, which correspond to regions of more uniform dose. These regions will be communicated to the user as good potential candidates for dosimeter placement for physical dose mapping experiments, because the low dose gradient minimizes the change of variability in experimental dose mapping due to human error in dosimeter placement. In an embodiment, a group of virtual tiles is identified as a potential candidate for locating a virtual or physical dosimeter, when that group constitutes a surface area of unmated surfaces in thevirtual representation of an object, and when that surface area is larger than a pre-specified threshold with a dose gradient magnitude lower than a pre-specified threshold.

[0077] The surface tiling and subsequent analysis of the surface dose distribution can be applied to each configuration in the family of configurations produced according to embodiments of the invention (e.g., using the method illustrated in Fig. 1).

[0078] In embodiments, the present invention also includes a computer interface (e.g., input device 1024 of Fig. 10) that allows a user to select a location on any surface of a display of the CAD model and be presented with a display of the dose delivered to the virtual tile nearest to the location that the user selected. For example, the virtual representation of the object, including the virtual tiles, may be displayed on a display device of a computer system (e.g., device 1026 of Fig. 10), and a dose received by a virtual tile and an uncertainty of the dose received by the virtual tile may then be displayed in response to a selection of the location in the display of the virtual representation of the object. The uncertainty of this dose can also be calculated by calculating the standard deviation over the dose of all neighboring tiles. This uncertainty calculation can account for the statistical uncertainty of the dose delivered to each surface tile that arises from the nature of radiation simulations. Additional information related to the location selected by the user may also be displayed, such as a part or component that underlies the surface tile selected, the material of the part / component, and its location.

[0079] Embodiments according to the present invention also include a novel method to calculate virtual dose maps for radiation (e.g., gamma) sterilization. This method models (simulates) incident radiation from a radiation source in a radiation processing (sterilization) facility that produces a given dose distribution in a product.

[0080] An example of a typical radiation processing facility is shown in Fig. 7. The product is loaded onto totes 700 and then travels through the irradiation room past a radiation (e.g., gamma) source 701. The totes move along a designated path 702, and stop (dwell) along the path at specific positions 703 (dwell locations) to accumulate the desired dose (this is known as the "shuffle-dwell" process). At each position, the totes may have a different distance and orientation relative to the radiation source.

[0081] Additionally, at many positions, the totes are “shielded” or blocked by other totes that are between them and the radiation source, which results in an attenuation of the radiation 704. This is a well-recognized occurrence in the radiation processing industry, and a scheduling process is used to produce a sequence of tote loadings that ensures that each product receives a dose within the prescribed range. The scheduling process assumes thateach tote has an average density, and totes are scheduled such that low and high density totes are combined in a way to achieve an overall consistent density.

[0082] The disclosed embodiments model the radiation produced by a radiation source as a “phase space” distribution, where “phase space” refers to the position, direction, and energy of each particle or photon produced by the radiation source. A representation of an object (a virtual object) and a virtual surface that encompasses the virtual object is accessed from computer system memory. More specifically, with reference to Fig. 8. in step 800, a virtual surface or set of surfaces near the tote that carries the object (product) is created and stored in computer system memory, from which it can be accessed. In an embodiment, this surface is a sphere 900 that encloses the tote 901 and hence the object, which is on the tote (see Fig. 9). In general, the virtual surface may be a closed, three-dimensional shape, such as a sphere or a rectangular prism, or it may be a set of planar or non-planar segments, which may or may not intersect with each other.

[0083] In an embodiment, the radiation source is a gamma source. In other embodiments, the radiation source is an electron beam source or an X-ray beam source, and the phase space on the virtual surface is determined accordingly.The phase space distribution delivered to the virtual surface is then calculated (step 801 of Fig. 8) using one of the methods described below. In embodiments, in step 801, the phase space distribution of radiation delivered to the virtual surface is calculated and stored in computer system memory.

[0084] In an embodiment, the phase space is represented as a multi-dimensional matrix with entries of energy bins, location bins (for example, bins of latitude and longitude), and directional bins (e.g., bins of directional cosines). Dimensionality can be reduced if entries are degenerate.

[0085] In an embodiment, the phase space distribution is generated through radiation simulations. In this embodiment, the entire shuffle-dwell process is simulated, and the phase space information 902 (Fig. 9) of each photon or particle produced by the radiation source that reaches the virtual surface is saved and used to determine the phase space distribution (for ease of discussion, a photon can be considered to be a particle).

[0086] In another embodiment, the phase space distribution is generated by an inverse process, instead of through direct simulations. Rather than determining what radiation is incident on outside of the virtual surface, this distribution is deduced from the dose that occurs to a product inside the virtual surface. This inverse process may use an iterative algorithm, where a candidate phase space distribution is created, a “forward” radiationsimulation is used to calculate the dose delivered to the product inside the virtual surface based on the candidate phase space distribution, and then the calculated dose to the product is compared against target dose data that is derived from other simulations or measurements. The difference between the calculated and target dose values can then be used to modify the phase space distribution, and the process is repeated iteratively until that difference satisfies a threshold value (e.g.. the difference reaches a minimum), at which point the phase space distribution is finalized.

[0087] An embodiment according to the invention uses the data generated through the qualification process that is laid out in the industry consensus standard ISO / ASTM 51702. Every sterilization facility has to conduct an Operational Qualification (OQ) by performing a sterilization run with a phantom comprising a three-dimensional grid of dosimeters 903. The measured dose delivered to this grid of dosimeters is used as the target dose described in the iterative algorithm above.

[0088] In another embodiment, a specially designed phantom with specified dosimeter locations is used to measure the dose inside the virtual surface.

[0089] In another embodiment, radiation simulations are used to simulate the dose delivered to a product within the virtual surface.

[0090] Once the phase space distribution is calculated, that phase space distribution can be stored in computer system memory and used as the source model to calculate the dose delivered to the object (product) inside the virtual surface (step 802). Accordingly, the dose distribution in any product inside the virtual surface can be modelled. Because the phase space distribution is accumulated over the dwell positions in the irradiation sequence, rather than simulating the entire shuffle-dwell process with many separate simulations, only one simulation is needed to calculate the dose distribution.

[0091] Figure 10 shows a block diagram of an example of a computer system 1000 upon which the embodiments described herein may be implemented. In its most basic configuration, the system 1000 includes at least one processing unit 1002 and memory 1004. This most basic configuration is illustrated in Figure 10 by dashed line 1006. The system 1000 may also have additional features and / or functionality. For example, the system 1000 may also include additional storage (removable and / or non-removable) including, but not limited to, magnetic or optical disks or tape. Such additional storage is illustrated in Figure 10 by removable storage 1008 and non-removable storage 1020. The system 1000 may also contain communications connect! on(s) 1022 that allow the device to communicate with otherdevices, e.g., in a networked environment using logical connections to one or more remote computers.

[0092] The system 1000 also includes input device(s) 1024 such as a keyboard, mouse, pen, voice input device, touch input device, etc. Output device(s) 1026 such as a display device, speakers, printer, etc., may also be included. A display device may be, for example, a cathode ray tube display, a light-emitting diode display, or a liquid crystal display.

[0093] In the example of Figure 10, the memory 1004 includes computer-readable instructions, data structures, program modules, and the like associated with implementing the methods for generating and modifying digital mechanical models to facilitate improved simulations of radiation processing applications that are disclosed herein.

[0094] The process parameters and sequence of steps described above are given by way of example only and can be varied as needed or desired. For example, while the steps illustrated and / or described above may be shown or discussed in a particular order, these steps do not necessarily need to be performed in the order illustrated or discussed. The various example methods described and / or illustrated herein may also omit one or more of the steps described herein or include additional steps in addition to those disclosed.

[0095] Although the subject matter has been described in language specific to structural features and / or methodological acts, it is to be understood that the subject matter defined in the disclosure is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the disclosure.

[0096] Embodiments according to the invention are thus described. While the present disclosure has been described in particular embodiments, the invention should not be construed as limited by such embodiments, but rather construed according to the following claims.

Claims

CLAIMSWhat is claimed is:

1. A computer-implemented method for radiation processing of an object, the method comprising: accessing a representation of the object, the representation comprising a plurality of virtual elements representing elements of the object; perturbing the representation to position and orient each virtual element of the plurality of virtual elements to generate a configuration of the virtual elements; and simulating radiation from a radiation source and calculating a dose received by said each virtual element in the configuration.

2. The computer-implemented method of claim 1, wherein the representation comprises a monolithic model comprising a plurality of component models integrated to form the monolithic model, wherein each component model of the plurality of component models comprises a respective representation of a respective virtual element of the plurality of virtual elements.

3. The computer-implemented method of claim 1, further comprising storing, in computer memory, geometric transformations that achieved, as a result of said perturbing, a position and orientation of said each virtual element in the configuration.

4. The computer-implemented method of claim 1. wherein said perturbing comprises automatically rearranging selected virtual elements of the plurality of virtual elements according to a specified pattern.

5. The computer-implemented method of claim 1, further comprising creating a first virtual element of the plurality of virtual elements based on a specified constraint, wherein the specified constraint is that the first virtual element has a void large enough to fully contain a second virtual element of the plurality of virtual elements.

6. The computer-implemented method of claim 1, wherein said perturbing comprises using physics-based modelling software to subject the virtual elements to modelled forces.

7. The computer-implemented method of claim 6, wherein the forces are derived from events that are expected to occur during handling and transportation of the object.

8. The computer-implemented method of claim 1, further comprising: producing a scanned volume of a physical realization of the object using an imaging method; and using the scanned volume to define the virtual representation of the object.

9. The computer-implemented method of claim 1, wherein said perturbing comprises applying an optimization algorithm to an optimization problem with specified constraints to vary positions and orientations of the virtual elements into the configuration, wherein the configuration satisfies the optimization problem.

10. The computer-implemented method of claim 1 , wherein said perturbing further comprises perturbing the representation to position and orient said each virtual to generate a plurality of configurations comprising different configurations of the virtual elements, and wherein said calculating further comprises: calculating a dose distribution for each of said different configurations; calculating locations and values of a minimum dose and a maximum dose of the dose distribution for said each different configuration; and calculating a dose uniformity ratio of the dose distribution for said each different configuration.

11. The computer-implemented method of claim 10, wherein said calculating the dose distribution comprises determining a mass density distribution based on the configuration, and wherein the dose distribution is modelled as a function dependent on the mass density distribution.

12. The computer-implemented method of claim 10, further comprising: calculating locations and values of a minimum dose and a maximum dose over all calculated dose distributions; and calculating a dose uniformity ratio over said all calculated dose distributions.

13. The computer-implemented method of claim 1, further comprising generating a plurality of virtual tiles that cover at least a portion of a surface of the virtual representation of the object.

14. The computer-implemented method of claim 1, further comprising calculating a statistical uncertainty in the dose.

15. The computer-implemented method of claim 1, wherein the radiation source is selected from the group consisting of: a gamma source, an electron beam source, and an X-ray source.

16. A computer-implemented method for radiation processing of an object, the method comprising: accessing a representation of the obj ect; generating a plurality of virtual tiles that are located on a surface of the representation of the object, wherein the virtual tiles are at locations on the surface corresponding to locations on the obj ect that are in contact with an environment surrounding the obj ect; and simulating radiation from a radiation source and calculating a dose received by each virtual tile of the plurality of virtual tiles.

17. The computer-implemented method of claim 16, wherein the virtual tiles have a configuration selected from the group consisting of: the virtual tiles are separated from each other by gaps; the virtual tiles overlap each other; the virtual tiles have different sizes and shapes; the virtual tiles have a tetrahedral shape.

18. The computer-implemented method of claim 16, further comprising calculating a statistical uncertainty in the dose.

19. The computer-implemented method of claim 16, wherein said calculating further comprises: calculating locations and values of a minimum tile dose and a maximum tile dose; and calculating a dose uniformity ratio.

20. The computer-implemented method of claim 16, further comprising: grouping a set of neighboring virtual tiles of the plurality of virtual tiles, the set comprising a virtual dosimeter; and calculating an average dose received by the virtual dosimeter.

21. The computer-implemented method of claim 16, further comprising calculating a spatial dose gradient for doses received by the plurality of virtual tiles.

22. The computer-implemented of claim 21, further comprising displaying the spatial dose gradient on a display device of a computer system.

23. The computer-implemented method of claim 16, further comprising identifying a group of virtual tiles of the plurality of virtual tiles that constitute a surface area in the representation of the object that is larger than a specified threshold with a dose gradient magnitude that is lower than a specified threshold, wherein the group of virtual tiles comprises a candidate for a location of a dosimeter.

24. The computer-implemented method of claim 16, further comprising displaying, on a display device of a computer system, a dose received by a virtual tile of the plurality of tiles and an uncertainty of the dose received by the virtual tile, wherein said displaying is in response to a selection of a location in a display of the representation of the object including the virtual tiles.

25. The computer-implemented method of claim 16, further comprising applying a spatial filter to a set of neighboring virtual tiles of the plurality of virtual tiles.

26. The computer-implemented method of claim 25, wherein the spatial filter is a smoothing filter.

27. A computer-implemented method for radiation processing of an object, the method comprising: accessing a representation comprising a virtual object representing the object and a virtual surface that encompasses the virtual obj ect; calculating a phase space distribution of radiation delivered to the virtual surface; and storing the phase space distribution in memory' of the computer.

28. The computer-implemented method of claim 27, further comprising simulating radiation from a radiation source using the phase space distribution to calculate a dose received by the virtual object.

29. The computer-implemented method of claim 28, wherein the radiation source is selected from the group consisting of: a gamma source, an electron beam source, and an X-ray source.

30. The computer-implemented method of claim 27, wherein said calculating the phase space distribution comprises representing a phase space as a multi-dimensional matrix with entries of energy bins, location bins, and directional bins.

31. The computer-implemented method of claim 27, wherein said calculating the phase space distribution comprises simulating the radiation delivered to the virtual surface from a radiation surface, and wherein the phase space distribution comprises phase space information of the radiation received by the virtual surface according to said simulating.

32. The computer-implemented method of claim 27, wherein said calculating the phase space distribution comprises: a) accessing a candidate phase space distribution; b) calculating a dose received by the virtual object based on the candidate phase space distribution; c) determining a difference between the dose and the target dose data; d) modifying the candidate phase space distribution according to the difference, to produce a modified phase space distribution; and e) using the modified phase space distribution as the phase space distribution if the difference satisfies a threshold value, and otherwise repeating the steps a) through e) using the modified phase space distribution as a next said candidate phase space distribution, wherein said repeating is performed until the difference satisfies the threshold value.

Images