Radiation therapy planning system and method

Abstract

The present specification provides a system for radiotherapy treatment planning. The system can obtain planning information related to at least one field to be applied to a subject in treatment. The system can also generate inputs to a flux map generation model based on the planning information. For each of the at least one field, the system can also generate at least one implementation flux map related to at least one sub-field of the field based on the input and the flux map generation model.

Description

Technical Field

[0001] The present description relates to radiation therapy (RT) and, more particularly, to a system and method for treatment planning in radiation therapy. Background Art

[0002] Radiation therapy (or radiotherapy) has been widely used in the clinical treatment of cancer and other conditions. Before radiation therapy, a treatment plan can be developed, which can define how the radiation therapy will be delivered to the subject. The accuracy of the treatment plan may affect the accuracy and effectiveness of the radiation therapy. Summary of the invention

[0003] One aspect of the specification provides a radiation therapy planning system, the system comprising: at least one storage device including a set of instruction sets; and at least one processor configured to communicate with the at least one storage device. When executing the instruction set, the at least one processor is configured to instruct the system to perform one or at least two of the following operations. The system can obtain planning information, which is related to at least one field to be applied to the object during treatment. The system can also generate inputs to a flux map generation model based on the planning information. For each of the at least one field, the system can also generate at least one implementation flux map associated with at least one sub-field of the field based on the input and the flux map generation model.

[0004] In some embodiments, the planning information includes segmentation information of one or more regions of interest (ROI) of the object to which the field is to be applied; and a field angle of each of the at least one field.

[0005] In some embodiments, the planning information also includes a reference image of the object.

[0006] In some embodiments, the planning information includes an optimal flux map for each of the at least one field.

[0007] In some embodiments, in order to obtain planning information related to at least one field to be applied to the subject in treatment, the system may perform the following operations: for each of the at least one field, obtain an initial flux map of the field; and generate an optimal flux map of the field by optimizing the initial flux map.

[0008] In some embodiments, for each of the at least one field, the field includes at least two sub-fields, and the at least one implemented flux map associated with the at least two sub-fields of the field includes a composite flux map of the at least two sub-fields.

[0009] In some embodiments, for each of the at least one field, the system may further convert the composite flux map into at least two sub-field flux maps, each of the sub-field flux maps corresponding to one of the at least two sub-fields.

[0010] In some embodiments, for each of the at least one field, the at least one implemented flux map associated with at least one sub-field of the field includes at least one sub-field flux map, each of the sub-field flux maps corresponding to one of the at least one sub-field.

[0011] In some embodiments, the flux map generation model includes at least one of a convolutional neural network (CNN) or a generative adversarial network (GAN).

[0012] In some embodiments, the flux map generation model is generated by the following training process. At least one training sample is obtained, wherein each training sample includes sample plan information and at least one implementation flux map true value, the sample plan information is related to at least one sample field to be applied to the sample object, and the at least one implementation flux map true value is related to at least one sample sub-field of the at least one sample field. The training process may also include generating the flux map generation model by training an initial model using the at least one training sample.

[0013] In some embodiments, for each of the at least one training sample, the acquiring of the training sample comprises one or more of the following operations. An initial sample flux map of the sample field may be acquired. An optimal sample flux map of the sample field may be generated by optimizing the initial sample flux map. The optimal sample flux map is converted into at least one initial sample sub-field flux map of at least one sample sub-field of the sample field. Based on the at least one initial sample sub-field flux map, at least one implementation flux map truth value associated with the at least one sample sub-field is generated.

[0014] According to one aspect of the present specification, a method for radiotherapy planning is provided. The method can be implemented on a computer device including at least one processor and at least one computer-readable storage medium for radiotherapy planning. The method includes obtaining planning information, the planning information being related to at least one field to be applied to the object during treatment. The method also includes generating inputs to a flux map generation model based on the planning information. For each of the at least one field, the method also includes generating at least one implementation flux map associated with at least one sub-field of the field based on the input and the flux map generation model.

[0015] According to one aspect of the present specification, a non-transitory computer-readable storage medium is provided. The non-transitory computer-readable storage medium includes a set of instructions for radiotherapy planning, and when executed by at least one processor, the instruction set guides the at least one processor to implement the following method. The method includes obtaining planning information, which is related to at least one field to be applied to the object during treatment. The method may also include generating inputs to a flux map generation model based on the planning information. For each of the at least one field, the method may also include generating at least one implementation flux map related to at least one sub-field of the field based on the input and the flux map generation model.

[0016] Some additional features of this specification may be explained in the following description. Some additional features of the present application will be apparent to those skilled in the art through study of the following description and corresponding drawings or understanding of the production or operation of the embodiments. The features and implementations of the present application may be realized and achieved by practicing or using various aspects of the methods, tools, and combinations set forth in the detailed examples discussed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] This specification will be further described in the form of exemplary embodiments, which will be described in detail by means of the accompanying drawings. The accompanying drawings are not drawn to scale. These embodiments are not restrictive, and in these embodiments, the same numbers represent the same structure, wherein:

[0018] Figure 1 is a schematic diagram of an exemplary RT planning system according to some embodiments of the present specification;

[0019] Figure 2 is a schematic diagram of exemplary hardware and / or software components of a computer device according to some embodiments of the present specification;

[0020] Figure 3 is a schematic diagram of exemplary hardware and / or software components of a mobile device according to some embodiments of the present specification;

[0021] Figure 4A and 4B is an exemplary structural diagram of a processing device according to some embodiments of this specification;

[0022] Figure 5 is a flowchart of an exemplary process for generating at least one implementation flux map according to some embodiments of the present specification.

[0023] Figure 6 is a schematic diagram of an exemplary flux map generation model according to some embodiments of this specification;

[0024] Figure 7 is a schematic diagram of an exemplary flux map generation model according to some embodiments of this specification;

[0025] Figure 8 is a schematic diagram of an exemplary flux map generation model according to some embodiments of this specification; and

[0026] Fig. 9 is a flowchart of an exemplary process for generating a flux map generation model according to some embodiments of the present specification;

[0027] Fig.10 is a flowchart of an exemplary process for generating at least one implementation flux map truth value of a training sample according to some embodiments of the present specification; and

[0028] Fig.11 It is a schematic diagram of a sub-field flux diagram of a sub-field of a radiation field according to some embodiments of the present specification. DETAILED DESCRIPTION

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. However, it should be understood by those skilled in the art that the present application can be implemented without these details. In other cases, in order to avoid unnecessarily obscuring the various aspects of the present application, well-known methods, processes, systems, components and / or circuits have been described at a higher level. For those of ordinary skill in the art, it is obvious that various changes can be made to the disclosed embodiments, and without departing from the principles and scope of the present application, the general principles defined in the present application can be applied to other embodiments and application scenarios. Therefore, the present application is not limited to the embodiments shown, but conforms to the broadest scope consistent with the scope of the application.

[0030] The terms used in this application are only for the purpose of describing specific example embodiments and are not restrictive. The singular forms "one", "an" and "the" used in this application may also include plural forms unless the context clearly indicates an exception. It should also be understood that the terms "including" and "comprising" used in this application specification only indicate the presence of the features, integers, steps, operations, components and / or parts, but do not exclude the presence or addition of one or more other features, integers, steps, operations, components, parts and / or combinations thereof.

[0031] It is to be understood that the terms "system", "engine", "unit", "module" and / or "block" used herein are methods for distinguishing different components, elements, parts, portions or assemblies at different levels in ascending order. However, these terms may also be replaced by other expressions if the same purpose can be achieved.

[0032] Generally, the terms "module", "unit" or "block" as used herein refer to logic embodied in hardware or firmware, or a collection of software instructions. The modules, units or blocks described herein may be implemented as software and / or hardware, and may be stored in any type of non-transitory computer-readable medium or another storage device. In some embodiments, software modules / units / blocks may be compiled and linked into an executable program. It should be understood that software modules may be called from other modules / units / blocks or from themselves, and / or may be called in response to detected events or interrupts. A computer program may be provided on a computer-readable medium for use in a computer device (e.g., Figure 2 210) executed on a computer readable medium such as a compact disc, a digital video disc, a flash drive, a magnetic device. A compact disc or any other tangible medium, or as a digital download (and may be initially stored in a compressed or installable format, requiring installation, decompression or decryption before execution). The software code here may be stored in part or in whole in a storage device of a computer device that performs the operation and applied to the operation of the computer device. Software instructions may be embedded in firmware such as an EPROM. It should also be understood that hardware modules / units / blocks may be included in connected logical components, such as gates and triggers, and / or may include programmable units, such as programmable gate arrays or processors. The modules / units / blocks or computer device functions described here may be implemented as software modules / units / blocks, but may be represented by hardware or firmware. Generally, the modules / units / blocks described here refer to logical modules / units / blocks, which may be combined with other modules / units / blocks or divided into sub-modules / sub-units / sub-blocks, although they are physical organizations or storage devices. This description may apply to a system, an engine, or a portion thereof.

[0033] It is understood that, unless the context clearly indicates otherwise, when a unit, engine, module, or block is referred to as being "on," "connected," or "coupled to" another unit, engine, module, or block, it may be directly on, connected, coupled, or in communication with the other unit, engine, module, or block, or there may be intermediate units, engines, modules, or blocks. In this application, the term "and / or" may include any one or more of the relevant listed items or combinations thereof.

[0034] These and other features, characteristics and functions and methods of operation of the related structural elements of the present application, as well as the combination of components and manufacturing economy, may become more apparent from the following description of the accompanying drawings, which constitute a part of the specification of this application. However, it should be understood that the drawings are for illustration and description purposes only and are not intended to limit the scope of the present application. It should be understood that the drawings are not drawn to scale.

[0035] Provided herein are systems and components for non-invasive imaging and / or treatment, for example, for disease diagnosis, treatment or research. In some embodiments, the system may include an RT system, a computed tomography (CT) system, an emission computed tomography (ECT) system, an X-ray imaging system, a positron emission tomography (PET) system, or the like, or any combination thereof. For illustrative purposes, the present application describes systems and methods for radiation therapy.

[0036] In the present application, the term "image" is used to collectively refer to image data (e.g., scan data, projection data) and / or various forms of images, including two-dimensional (2D) images, three-dimensional (3D) images, four-dimensional (4D) images, etc. The terms "pixel" and "voxel" in the present application are used interchangeably to refer to elements of an image. The term "anatomical structure" in the present application may refer to a gas (e.g., air), liquid (e.g., water), solid (e.g., stone), cell, tissue, organ, or any combination thereof of an object, which may be displayed in an image (e.g., planning imaging or treatment imaging, etc.) and actually exists in or on the body of a subject. The terms "part", "position", and "region" in the present application may refer to the location of an anatomical structure shown in an image or the actual location of an anatomical structure existing in or on a subject, because an image may indicate the actual location of certain anatomical structures existing in or on a subject.

[0037] Radiation therapy has been widely used in the clinical treatment of cancer and other diseases. Treatment planning is an important part of radiation therapy, and the accuracy of the treatment plan generated during the treatment planning process may affect the treatment effect and / or treatment accuracy. Typically, during radiation therapy of a subject (e.g., a cancer patient), one or more fields may be applied to the subject to treat the subject, wherein each field includes at least two sub-fields. During the treatment planning stage, one or more implementation flux maps may need to be generated to guide how to apply the fields and / or sub-fields of the fields.

[0038] Traditionally, treatment planning techniques can generate an initial flux map of a field, optimize the initial flux map to generate an optimal flux map of the field, and further divide the optimal flux map of the field into one or more implementation flux maps. The division of the flux map of the field usually involves generating at least two initial sub-field flux maps, and an iterative or manual optimization process of the initial sub-field flux maps. Traditional treatment planning techniques may require a large amount of computing resources, resulting in a long treatment planning time, and / or be susceptible to human error or subjectivity.

[0039] One aspect of the specification relates to systems and methods for treatment planning in radiation therapy. The systems and methods can obtain planning information related to at least one field to be applied to an object in treatment. The systems and methods generate inputs to a flux map generation model based on the planning information. For each of the at least one field, the systems and methods can generate at least one implementation flux map associated with at least one sub-field of the field based on the input and the flux map generation model. For example, at least one implementation flux map of the field can include a composite flux map of at least two sub-fields of the field, a sub-field flux map of each sub-field of the field, and the like, or any combination thereof. At least one implementation flux map of the field can be used to guide the application of the field during treatment of the object.

[0040] Compared with traditional treatment planning techniques, the systems and methods of the present specification can efficiently and / or accurately generate at least one implementation flux map of a field. For example, a flux map generation model can be applied to treatment planning to learn the optimal mechanism for generating an implementation flux map from training data. The application of the flux map generation model can avoid generating an initial flux map of the field, optimizing the initial flux map, and / or dividing the optimal flux map, which can improve the efficiency of the treatment plan, for example, reducing the user's workload, the impact of user differences, and / or the time required for treatment planning. In addition, in some embodiments of the present specification, the generation of the implementation flux map of the field can be done with little or no direct human intervention, which is more objective and reliable, not affected by human error or subjectivity, and / or fully automated.

[0041] Figure 1 is a schematic diagram of an exemplary RT planning system 100 according to some embodiments of the present specification. The RT planning system 100 may include a radiation delivery device 110, a network 120, one or more terminals 130, a processing device 140, and a storage device 150. In some embodiments, two or more components of the RT planning system 100 may be connected to each other and / or communicate with each other via a wireless connection (e.g., the network 120), a wired connection, or a combination thereof. The connections between the components of the RT planning system 100 may vary. By way of example only, the radiation delivery device 110 may be connected to the processing device 140 via the network 120 or directly. As another example, the storage device 150 may be connected to the processing device 140 via the network 120 or directly. As yet another example, the terminal 130 may be connected to the processing device 140 directly or via the network 120.

[0042] In some embodiments, the radiation delivery device 110 can be an RT device. The RT device can be configured to provide radiation therapy for cancer and other patients. For example, the RT device can apply one or more radiation fields to a treatment area (e.g., a tumor) of a subject (e.g., a patient) to relieve the subject's symptoms. In some embodiments, the RT device can be a combined radiation therapy device, such as an image-guided radiation therapy (IGRT) device, an intensity-modulated radiation therapy (IMRT) device, an arc-shaped intensity-modulated radiation therapy (IMAT) device, etc.

[0043] like Figure 1 As shown, in some embodiments, the radiation delivery device 110 may include an imaging component 113, a treatment component 116, a bed 114, etc. The imaging component 113 can be used to obtain images of the object during radiation therapy staging and / or after radiation treatment. The object may include any biological object (e.g., a human, an animal, a plant, or a part thereof) and / or a non-biological object (e.g., a phantom). For example, the imaging component 113 may include a computed tomography (CT) device (e.g., a cone beam computed tomography (CBCT) device, a fan beam computed tomography (FBCT) device), an ultrasound imaging device, a fluoroscopic imaging device, a magnetic resonance imaging (MRI) device, a single photon emission computed tomography (SPECT) device, a positron emission tomography (PET) device, an X-ray imaging device, etc. or any combination thereof.

[0044] In some embodiments, the imaging component 113 may include an imaging radiation source 115, a detector 112, a rack 111, etc. The imaging radiation source 115 and the detector 112 may be mounted on the rack 111. The imaging radiation source 115 may emit radiation to the object. The detector 112 may detect radiation (e.g., X-ray photons, gamma ray photons) emitted from the imaging area of ​​the imaging component 113. In some embodiments, the detector 112 may include one or more detection units. The detector unit may include a detector (e.g., a detector, an oxychloryl sulfide detector), a gas detector, etc. The detector unit may include a single-row detector and / or a multi-row detector.

[0045] The treatment component 116 may be configured to apply radiation therapy to the object. The treatment component 116 may include a treatment head and a gantry 118. In some embodiments, the treatment component 116 may include a treatment radiation source 117, a collimator 119, and the like. The treatment radiation source 117 may be configured to generate and emit radiation for treatment to the object. The collimator 119 may be configured to control the shape of the radiation beam generated by the treatment radiation source 117. In some embodiments, the gantry 118 may be rotatable, and the rotation of the gantry 118 may drive the treatment head to rotate around the object. During the rotation of the gantry 118, the treatment radiation source 117 may emit radiation toward the object at different field angles. In some embodiments, when the gantry 118 is located at a specific angle, the treatment radiation source 117 may deliver at least two sub-fields to the object, which have different shapes formed by the collimator 119. In some embodiments, the radiation beam emitted by the treatment radiation source 117 may include electrons, photons, or other types of radiation. In some embodiments, the energy of the radiation beam may be in the megavolt range (e.g., >1 MeV), which may be referred to as a megavolt (MV) field. In some embodiments, treatment radiation source 117 may include a linear accelerator (LINAC) for accelerating electrons, ions, or protons.

[0046] In some embodiments, the imaging component 113 can be spaced a certain distance from the treatment component 116. In some embodiments, the frame 111 of the imaging component 113 of the treatment component 116 and the frame 118 of the treatment component 116 can share a rotation axis. The object can be placed on the bed 114 for treatment and / or imaging. In some embodiments, the imaging radiation source 115 and the treatment radiation source 117 can be integrated into one radiation source to image and / or treat the object. In some embodiments, the imaging component 113 and the treatment component 116 can share the same frame. For example, the treatment radiation source 117 can be mounted on the frame 111 of the imaging component 113.

[0047] The network 120 may include any suitable network that can be used for information and / or data exchange of the RT planning system 100. In some embodiments, one or more components of the RT planning system 100 (e.g., the radiation delivery device 110, the terminal 130, the processing device 140, the storage device 150, etc.) may communicate information and / or data with one or more other components of the RT planning system 100 via the network 120. For example, the processing device 140 may obtain image data from the radiation delivery device 110 via the network 120. As another example, the processing device 140 may obtain user instructions from the terminal 130 via the network 120. The network 120 may be or include a public network (e.g., the Internet), a private network (e.g., a local area network (LAN)), a wired network, a wireless network (e.g., an 802.11 network, a Wi-Fi network), a frame relay network, a virtual private network (VPN), a satellite network, a telephone network, a router, a hub, a switch, a server computer, and / or any combination thereof. For example, the network 120 may include a cable network, a wired network, an optical fiber network, a telecommunication network, an intranet, a wireless local area network (WLAN), a metropolitan area network (MAN), a public switched telephone network (PSTN), a Bluetooth TM Network, ZigBee TM Network, near field communication (NFC) network, etc., or any combination thereof. In some embodiments, the network 120 may include one or more network access points. For example, the network 120 may include a wired and / or wireless network access point such as a base station and / or an Internet exchange point, and one or more components of the RT planning system 100 may be connected to the network 120 through the wired and / or wireless network access point to exchange data and / or information.

[0048] The terminal 130 may be used to implement interaction between the user and the RT planning system 100. In some embodiments, the terminal 130 may include a mobile device 131, a tablet computer 132, a laptop computer 133, etc. or any combination thereof. In some embodiments, the mobile device 131 may include a smart home device, a wearable device, a smart mobile device, a virtual reality device, an augmented reality device, etc. or any combination thereof. For example only, the terminal 130 may include Figure 3The mobile device shown. In some embodiments, smart home devices may include smart lighting devices, smart appliance control devices, smart monitoring devices, smart TVs, smart cameras, intercoms, etc., or any combination thereof. In some embodiments, wearable devices may include bracelets, footwear, glasses, helmets, watches, clothing, backpacks, smart accessories, etc., or any combination thereof. In some embodiments, mobile devices may include mobile phones, personal digital assistants (PDAs), gaming devices, navigation devices, point-of-sale (POS) devices, laptop computers, tablet computers, desktop computers, etc., or any combination thereof. In some embodiments, virtual reality devices and / or augmented reality devices may include virtual reality helmets, virtual reality glasses, virtual reality goggles, augmented reality helmets, augmented reality glasses, augmented reality goggles, etc., or any combination thereof. For example, virtual reality devices and / or augmented reality devices may include Google Glass. TM 、Oculus Rift TM 、Hololens TM 、GearVR TM Etc. In some embodiments, the terminal 130 may be a part of the processing device 140 .

[0049] The processing device 140 may process information obtained from the radiation delivery device 110, the terminal 130, and / or the storage device 150. For example, the processing device 140 may generate at least one implementation flux map through a flux map generation model. As another example, the processing device 140 may generate a flux map generation model by training an initial model using a plurality of training samples. In some embodiments, the processing device may perform generation and / or update of the flux map generation model, and the application of the flux map generation model may be performed by another processing device. In some embodiments, the flux map generation model may be executed by a system different from the RT planning system 100 or a server different from the processing device 140. For example, the flux map generation model may be generated by a first system of a supplier that provides and / or maintains the model, and / or has a training sample for obtaining the model, and the flux map generation based on the flux map generation model may be performed by a second system of a customer of the supplier. In some embodiments, the flux map generation model may be generated online in response to a request for flux map generation. In some embodiments, the flux map generation model may be generated offline.

[0050] In some embodiments, the flux map generation model may be generated and / or updated (or maintained) by, for example, a manufacturer or supplier of the radiation delivery device 110. For example, the manufacturer or supplier may load the flux map generation model into the RT planning system 100 or a portion thereof (e.g., the processing device 140) before or during installation of the radiation delivery device 110 and / or the processing device 140, and maintain or update the flux map generation model in real time (regularly or irregularly). The maintenance or update may be achieved by an installation program, which may be stored on a storage device (e.g., a CD, a USB drive, etc.) or searched from an external source (e.g., a server maintained by a manufacturer or supplier) using the network 120. The program may include a new model (e.g., a new flux map generation model) or a portion of a model that replaces or supplements a corresponding portion of the original flux map generation model.

[0051] In some embodiments, the processing device 140 may be a computer, a user console, a single server, a server group, etc. The server group may be centralized or distributed. In some embodiments, the processing device 140 may be local or remote. For example, the processing device 140 may access information stored in the radiation delivery device 110, the terminal 130, and / or the storage device 150 via the network 120. For another example, the processing device 140 may be directly connected to the radiation emission device 110, the terminal 130, and / or the storage device 150 to access the stored information. In some embodiments, the processing device 140 may be implemented on a cloud platform. By way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, the like, or any combination thereof. In some embodiments, the processing device 140 may be provided by a computer having a plurality of computer systems such as the computer system 100, ... Figure 2 One or more of the components shown may be implemented by a computer device 200 .

[0052] The storage device 150 may store data, instructions, and / or any other information. In some embodiments, the storage device 150 may store data obtained from the terminal 130 and / or the processing device 140. In some embodiments, the storage device 150 may store data and / or instructions used by the processing device 140 to perform the exemplary methods described in the present application. In some embodiments, the storage device 150 may include a mass storage device, a removable storage device, a volatile read-write memory, a read-only memory (ROM), etc., or any combination thereof. Exemplary mass storage devices may include disks, optical disks, solid-state drives, etc. Exemplary removable storage devices may include flash drives, floppy disks, optical disks, memory cards, zip disks, tapes, etc. Exemplary volatile read-write memory may include random access memory (RAM). Exemplary RAM may include dynamic random access memory (DRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), static random access memory (SRAM), thyristor random access memory (T-RAM), and zero capacitance random access memory (Z-RAM), etc. Exemplary ROMs may include mask ROM (MROM), programmable ROM (PROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), compact disk ROM (CD-ROM), and digital versatile disk ROM, etc. In some embodiments, the storage device 150 may be implemented on a cloud platform. By way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, etc., or any combination thereof.

[0053] In some embodiments, the storage device 150 may be connected to the network 120 to communicate with one or more other components of the RT planning system 100 (e.g., the processing device 140, the terminal 130). One or more components of the RT planning system 100 may access data and / or instructions stored in the storage device 150 through the network 120. In some embodiments, the storage device 150 may be directly connected to one or more other components of the RT planning system 100 (e.g., the processing device 140, the terminal 130). In some embodiments, the storage device 150 may be part of the processing device 140. In some embodiments, the storage device 150 may be connected to the radiation delivery device 110 or connected to the back end of the processing device 140 or communicate with the radiation delivery device 110 via the network 120.

[0054] It should be noted that the above description of the RT planning system 100 is intended to be illustrative and not to limit the scope of the present specification. Many alternatives, modifications, and variations will be apparent to those skilled in the art. The features, structures, methods, and other features of the exemplary embodiments described herein may be combined in various ways to obtain additional and / or alternative exemplary embodiments. For example, the RT planning system 100 may include one or more additional components. In addition, one or more components of the above-described RT planning system 100 (e.g., the imaging component 113 of the radiation delivery device 110) may be omitted. For another example, two or more components of the RT planning system 100 may be integrated into a single component.

[0055] Figure 2 2 is a schematic diagram of exemplary hardware and / or software components of a computer device 200 according to some embodiments of the present specification. The computer device 200 may be used to implement any component of the RT planning system 100 as described herein. For example, the processing device 140 and / or the terminal 130 may be implemented on the computer device 200 via their hardware, software programs, firmware, or a combination thereof, respectively. Although only one computer device is shown, for convenience, the computer functions related to the RT planning system 100 described herein may be distributed across multiple similar platforms to distribute the processing load. Figure 2 As shown, computer device 200 may include processor 210 , memory 220 , input / output (I / O) 230 , and communication port 240 .

[0056] The processor 210 can execute computer instructions (e.g., program code) and perform functions of the processing device 140 according to the techniques described herein. The computer instructions may include, for example, routines, programs, objects, components, data structures, processes, modules, and functions that perform specific functions described herein. For example, the processor 210 can process image data obtained from the radiation delivery device 110, the terminal 130, the storage device 150, and / or any other component of the RT planning system 100. In some embodiments, the processor 210 may include one or more hardware processors, such as a microcontroller, a microprocessor, a reduced instruction set computer (RISC), an application specific integrated circuit (ASIC), an application specific instruction set processor (ASIP), a central processing unit (CPU), a graphics processing unit (GPU), a physical processing unit (PPU), a microcontroller unit, a digital signal processor (DSP), a field programmable gate array (FPGA), an advanced RISC machine (ARM), a programmable logic device (PLD), any circuit or processor capable of performing one or more functions, etc., or any combination thereof.

[0057] For illustration only, only one processor is described in the computer device 200. However, it should be noted that the computer device 200 in the present application may also include multiple processors, and therefore, the operations and / or methods performed by one processor as described in the present application may also be performed jointly or separately by multiple processors. For example, if in the present application, the processor of the computer device 200 performs operation A and operation B at the same time, it should be understood that operation A and operation B may also be performed jointly or separately by two or more different processors in the computer device 200. (For example, the first processor performs operation A, the second processor performs operation B, or the first processor and the second processor perform operations A and B together).

[0058] The memory 220 may store data obtained from one or more components of the RT planning system 100. In some embodiments, the memory 220 may include a mass storage device, a removable storage device, a volatile read-write memory, a read-only memory (ROM), etc., or any combination thereof. In some embodiments, the memory 220 may store one or more programs and / or instructions to perform the exemplary methods described in the present application. For example, the memory 220 may store a program that is executed on the processing device 140 to generate at least one implementation flux map.

[0059] I / O 230 can input and / or output signals, data, information, etc. In some embodiments, I / O 230 can enable a user to interact with processing device 140. In some embodiments, I / O 230 can include input devices and output devices. Input devices can include alphanumeric and other keys, which can be input through a keyboard, a touch screen (e.g., with tactile or haptic feedback), voice input, eye tracking input, a brain monitoring system, or any other comparable input mechanism. Input information received by the input device can be sent to another component (e.g., processing device 140) via, for example, a bus for further processing. Other types of input devices can include cursor control devices, such as a mouse, a trackball, or cursor direction keys, etc. Output devices can include displays (e.g., liquid crystal displays (LCDs), displays based on light emitting diodes (LEDs), flat panel displays, curved screens, television devices, cathode ray tubes (CRTs) (touch screens), speakers, printers, etc., or combinations thereof.

[0060] The communication port 240 can be connected to a network (e.g., network 120) to facilitate data communication. The communication port 240 can establish a connection between the processing device 140 and the ray emitting device 110, the terminal 130 and / or the storage device 150. The connection can be a wired connection, a wireless connection, any other communication connection that can achieve data transmission and / or reception, and / or any combination of these connections. The wired connection can include, for example, an electrical cable, an optical cable, a telephone line, etc., or any combination thereof. The wireless connection can include, for example, Bluetooth TM Link, Wi-Fi TM Link, WiMax TM In some embodiments, the communication port 240 may include a wireless communication port, a WLAN link, a ZigBeeTM link, a mobile network link (e.g., 3G, 4G, 5G), etc., or a combination thereof. In some embodiments, the communication port 240 may be and / or include a standardized communication port, such as RS232, RS485, etc. In some embodiments, the communication port 240 may be a specially designed communication port. For example, the communication port 240 may be designed according to the Digital Imaging and Communications in Medicine (DICOM) protocol.

[0061] Figure 3 300 is a schematic diagram of exemplary hardware and / or software components of a mobile device according to some embodiments of the present specification. In some embodiments, the terminal 130 and / or the processing device 140 can be implemented on the mobile device 300, respectively. Figure 3 As shown, mobile device 300 may include communication platform 310, display 320, graphics processing unit (GPU) 330, central processing unit (CPU) 340, I / O 350, memory 360 and storage 390. In some embodiments, any other suitable components, including but not limited to a system bus or controller (not shown), may also be included in mobile device 300. In some embodiments, mobile operating system 370 (e.g., iOS 11) may be used. TM 、Android TM 、WindowsPhone TM ) and one or more applications 380 are loaded from storage 390 into memory 360 for execution by CPU 340. Application 380 may include a browser or any other suitable mobile application for receiving and presenting information related to RT planning system 100. User interaction with the information stream may be achieved through I / O 350 and provided to processing device 140 and / or other components of RT planning system 100 through network 120.

[0062] In order to implement the various modules, units and their functions described in this application, a computer hardware platform can be used as a hardware platform for one or more components described herein. A computer with user interface elements can be used to implement a personal computer (PC) or any other type of workstation or terminal device. If the computer is properly programmed, the computer can also be used as a server.

[0063] Figure 4A and 4B is an exemplary structural diagram of processing devices 140A and 140B according to some embodiments of this specification. In some embodiments, processing devices 140A and 140B may be as follows Figure 1 An example of the processing device 140 described herein. The processing device 140A may be configured to generate one or more implementation flux maps for the object to be treated. For example, the processing device 140A may apply a flux map generation model to generate one or more implementation flux maps. The processing device 140B may be configured to generate a flux map generation model through model training.

[0064] In some embodiments, processing devices 140A and 140B may each be in a processing unit (eg, Figure 2 The processor 210 shown or Figure 3 340). As an example only, processing device 140A may be implemented on the CPU 340 of the terminal device, and processing device 140B may be implemented on computer device 200. As another example, processing device 140A may be implemented on a computer device of RT planning system 100, and processing device 140B may be a device or system of a manufacturer of RT planning system 100 or a portion thereof (e.g., radiation delivery device 110). Alternatively, processing devices 140A and 140B may be implemented on the same computer device 200 or the same CPU 340. For example, processing devices 140A and 140B may be implemented on the same computer device 200.

[0065] like Figure 4A As shown, the processing device 140A may include an acquisition module 401 and a generation module 402. The acquisition module 401 may be configured to acquire information related to the RT planning system 100. For example, the acquisition module 401 may acquire planning information related to at least one field to be applied to the subject (e.g., a particular treatment stage) during treatment. More descriptions of acquiring planning information may be found elsewhere in this specification. For example, see step 510 and its related description.

[0066] The generation module 402 can be configured to generate inputs to a flux map generation model based on the plan information. As used herein, a flux map generation model refers to a model (e.g., a machine learning model) or an algorithm that generates an implementation flux map based on its inputs. More descriptions of the generation of the flux map generation model can be found elsewhere in this specification. For example, see step 520 and its related description. In addition, the generation module 402 can be configured to generate at least one implementation flux map related to at least one field based on the input and the flux map generation model. More descriptions of the generation of at least one implementation flux map can be found elsewhere in this specification. For example, see step 530 and its related description.

[0067] like Figure 4B As shown, the processing device 140B may include an acquisition module 403 , a training module 404 , and a generation module 405 .

[0068] The acquisition module 403 may be configured to acquire information related to the training of the flux map generation model. For example, the acquisition module 403 may acquire at least one training sample, each of which may include sample plan information and at least one implementation flux map truth value. The sample plan information of the training sample may involve at least one sample field to be applied to the sample object. More descriptions about acquiring training samples may be found elsewhere in this specification. For example, see step 910 and its related description.

[0069] The training module 404 may be configured to generate a flux map generation model by training an initial model using at least one training sample. More descriptions of the generation of the flux map generation model may be found elsewhere in this specification. For example, see step 920 and its related description.

[0070] The generation module 405 may be configured to generate at least one implementation flux map truth value of the training sample. For example, the generation module may perform the following steps: Fig.10 One or more steps of process 1000 are described to generate at least one implementation flux map ground truth for training samples.

[0071] It should be noted that the above description is provided for the purpose of illustration only and is not intended to limit the scope of the present application. For those of ordinary skill in the art, various changes and modifications can be made according to the description of this specification. However, these changes and modifications do not depart from the scope of the present application. In some embodiments, processing device 140A and processing device 140B can share two or more modules, and any module can be divided into two or more units. For example, processing device 140A and 140B can share the same acquisition module, that is, acquisition module 401 and acquisition module 403 are the same module. In some embodiments, processing device 140A and / or processing device 140B may include one or more additional modules, such as a storage module (not shown) for storing data. In some embodiments, processing device 140A and processing device 140B may be integrated into one processing device 140.

[0072] Figure 5 It is a flowchart of an exemplary process for generating at least one implementation flux map according to some embodiments of the present specification. In some embodiments, process 500 can be performed by RT planning system 100. For example, process 500 can be implemented by an instruction set (e.g., an application) stored in a storage device (e.g., storage device 150, storage device 220, and / or storage device 390). In some embodiments, processing device 140 (e.g., processor 210 of computer device 200, CPU 340 of mobile device 300, and / or one or more modules shown in FIG. 4) can execute the set of instruction sets, and when executing the instructions, processing device 140A can be configured to execute process 500. The operations of the process shown below are for illustrative purposes only. In some embodiments, process 500 can complete process 500 by one or more operations not described and / or omitting one or more operations in the operations discussed below. In addition, Figure 5 The order of the operations of flow 500 shown in FIG. 5 and described below is not intended to be limiting.

[0073] Generally, before radiation treatment is performed on a treatment object (e.g., a few days or weeks before), a planning image of the object can be acquired by an imaging device (e.g., a CT device, an MRI device), and a treatment plan can be generated based on the planning image. The treatment plan can describe how to perform radiation treatment on the object. As an example only, radiation treatment can be applied to the object in multiple treatment stages, and this process will be implemented within a treatment period lasting multiple days (e.g., 2 to 5 weeks). The treatment plan can include information, such as how one or more radiation fields are applied to the target area of ​​the object in each treatment stage during the treatment process. For example, the treatment plan can include the total dose of each treatment stage (e.g., 0.1Gy, 10Gg, 50Gg, 100Gg, etc.) and the dose distribution of the object.

[0074] In some embodiments, during treatment of an object (e.g., treatment staging), at least one field may be applied to the object from a specific field angle (or treatment angle) by a radiotherapy device to treat the object. Each field may be applied to the object by applying at least one sub-field to the object. For example, at least two sub-fields (or sub-fields) may be applied to the object in sequence. The at least two sub-fields corresponding to the field may be shaped into different shapes by a collimator (e.g., a multi-leaf collimator (MLC)). During the treatment planning stage, an initial flux map (or a planned flux map) corresponding to each field may be generated. The initial flux map of the field may display the planned distribution of the radiation dose applied to the object through all the sub-fields of the field. For illustrative purposes, the following describes how to apply a field through at least two sub-fields. It should be understood that this is not restrictive, and the field may include only one sub-field.

[0075] Conventionally, it may be necessary to optimize an initial flux map of a field to generate an optimal flux map of the field. For example, a flux map optimization algorithm may be used to optimize the initial flux map to generate an optimal flux map with higher quality (e.g., higher resolution, lower radiation dose to healthy organs, and / or more uniform dose distribution in the target area of ​​the object). Exemplary flux map optimization algorithms may include linear / nonlinear programming algorithms, mixed integer programming algorithms, simulated annealing algorithms, genetic algorithms, flux map optimization (FMO) algorithms, direct subfield optimization (DAO) algorithms, direct machine parameter optimization (DMPO) algorithms, and the like.

[0076] Then, the optimal flux map of the radiation field can be divided into at least two sub-fields (or referred to as applicable sub-fields), for example, using a leaf sequencing algorithm. For each sub-field, the initial sub-field flux map can be used to describe the planned distribution of the radiation dose delivered to the object through the sub-field. However, affected by the accuracy of the generated optimal flux map and / or the initial sub-field flux map, there may be an error between the initial composite flux map of the initial sub-field flux maps of at least two sub-fields and the optimal flux map. The error between the optimal flux map and the initial composite flux map may affect the accuracy of the treatment plan. Conventional treatment planning techniques may need to optimize the initial sub-field flux map of the sub-field to generate an optimal sub-field flux map that can be used to guide the application of the sub-field. For example, the initial sub-field flux map can be iteratively optimized so that the optimal composite flux map of the optimal sub-field flux map can match the optimal flux map. As another example, a user (e.g., a doctor, a radiologist) can manually adjust the parameters of the sub-field and / or the initial sub-field flux map in the interface of the terminal device. Optimization of the initial sub-field flux map may require significant computational resources, result in long treatment times, and / or be susceptible to human error or subjectivity.

[0077] It is necessary to provide a system and method for generating at least one implementation flux map of at least one field applied to an object. As used herein, an implementation flux map of a field refers to a flux map that can be used to guide the application of the field in radiation treatment. For example, at least one implementation flux map of a field may include at least two sub-field flux maps (or referred to as the optimal sub-field flux map described above), each sub-field flux map corresponding to one of the at least two sub-fields of the field. For another example, the at least one implementation flux map may include a composite flux map of the at least two sub-fields (or referred to as the optimal composite flux map described above).

[0078] For example, the process 500 described below can generate at least one implementation flux map for applying a field to an object through a flux map generation model. The flux map generation model refers to a model (e.g., a machine learning model) or algorithm that generates an implementation flux map based on its input. Traditional methods require generating at least two initial sub-field flux maps of the field and iteratively or manually optimizing the initial sub-field flux maps. Compared with this, the systems and methods disclosed herein can be more efficient and accurate, for example, avoiding the optimization of sub-field flux maps, reducing the workload of users, the impact of user differences, and the time required to generate at least one implementation flux map.

[0079] As used herein, a flux map of a field or one or more subfields of a field may be represented, for example, in the form of a graph, a chart, a table, etc. For illustrative purposes, Fig.11 FIG. 1 is a schematic diagram of a sub-field flux diagram 1130 of a sub-field 1120 of a radiation field according to some embodiments of the present specification. The sub-field 1120 of the radiation field can be adjusted by the MLC 1110 of the radiotherapy device. Fig.11 As shown, the sub-field flux map 1130 can be represented in a two-dimensional graph, which represents the distribution of radiation dose on a plane perpendicular to the incident direction of the sub-field 1120 (for example, the direction of the central axis of the field). The two-dimensional graph may include an X-axis corresponding to the horizontal direction of the plane and a Y-axis corresponding to the vertical direction of the plane. The points in the sub-field flux map 1130 may correspond to physical points or areas on the plane. The x-axis and y-axis coordinates of the points in the sub-field flux map 1130 may reflect the position of the corresponding physical point or area in the horizontal direction and the vertical direction of the plane, respectively. The value of the point in the sub-field flux map 1130 may reflect the intensity of photons irradiated on the corresponding physical point or area (that is, the radiation dose at the physical point or area). It should be noted that the provided Fig.11 The sub-field flux map 1130 shown in FIG. 1 is for illustrative purposes and is not intended to be limiting. The form of the sub-field flux map 1130 may be modified according to actual needs. For example, different points in the sub-field flux map 1130 may be displayed in different colors according to the value of the point.

[0080] At 510 , the processing device 140A (eg, the acquisition module 401 ) may acquire planning information related to at least one field to be applied to a subject during treatment (eg, a particular treatment phase).

[0081] The object may include a patient, a part of a patient, or any living being that needs to be treated by a radiotherapy device (e.g., the radiation delivery device 110). The planning information may include any information related to at least one field to be applied to the object. For example, the planning information may include characteristic information of the object (e.g., gender, age, height, width, thickness, etc.), characteristic information of one or more regions of interest (ROI) of the object, field angles of each of the at least one field, a reference image of the object, an initial flux map of each of the at least one field, an optimal flux map of each of the at least one field, a dose constraint of at least one field, a prescription for the object specified by a user (e.g., a doctor, a radiologist), etc., or any combination thereof.

[0082] The ROI of the object may include a target area and / or an organ at risk (OAR) near the target. The target area may include an area of ​​the object that includes at least a portion of malignant tissue (e.g., a tumor, a cancerous organ, or a non-cancerous target for radiotherapy). For example, the target area may be a tumor that needs to be treated by radiotherapy, an organ with a tumor, a tissue with a tumor, or any combination thereof. The OAR may include an organ and / or tissue near the target area that does not need to be irradiated but is at risk of being damaged by radiation due to being close to the target area. Exemplary feature information of the ROI of the object may include the position, contour, shape, height, width, thickness, area, ratio of height to width, etc. of the ROI, or any combination thereof.

[0083] In some embodiments, the feature information of the ROI may include segmentation information (or contour information or edge information) of the ROI. For example, the segmentation information of the ROI may include the contour of the ROI segmented from the reference image (or other image) of the object, one or more parameters describing the contour of the ROI (e.g., shape, height, width, thickness, area, ratio of height to width, etc.), or any combination thereof. More descriptions about segmentation information can be found elsewhere in this specification. For example, see Figure 6 and its related description.

[0084] The field angle of a field refers to the angle or direction of a sub-field of a field applied to an object during radiotherapy. The reference image of the object may be a two-dimensional (2D) image, a three-dimensional (3D) image (e.g., a time series of 3D images), a four-dimensional (4D) image, etc., or a combination thereof. The reference image of the object may include a CT image (e.g., a cone field CT (CBCT) image, a fan field CT (FBCT) image), an MR image, a PET image, an X-ray image, a fluoroscopic examination image, an ultrasound image, a radiotherapy radiation image, a SPECT image, etc., or a combination thereof. In some embodiments, the reference image of the object (e.g., a CT image) may be a planning image for formulating a treatment plan. The dose constraints associated with the at least one field may include, for example, a maximum total radiation dose applied to an object or a portion thereof (e.g., a target volume, an OAR) by at least one field, a maximum total radiation dose applied to an object (or a portion thereof) by each field, a maximum total radiation dose applied to an object (or a portion thereof) by each sub-field of each field, or any combination thereof. In some embodiments, the total radiation dose applied to an object by all at least one field may need to meet a specific constraint (e.g., be below a threshold). The subject's prescription may be provided by a user (eg, a physician or radiologist), which may include tumor location, tumor grade, maximum total radiation dose, location of radiation zones, or other information provided by the user, or any combination thereof.

[0085] In some embodiments, the planning information (or a portion thereof) may be pre-generated and stored in a storage device (e.g., the storage device 150, the storage device 220, or an external storage device). The processing device 140A may obtain the planning information (or a portion thereof) from the storage device. For example, the reference image of the object may be a historical medical image of the object stored in the storage device, and the processing device 140A may retrieve the historical medical image from the storage device as the reference image. Alternatively, the processing device 140A may cause an imaging device (e.g., the imaging component 113) to obtain a reference image by scanning the object, and obtain the reference image of the object from the imaging device.

[0086] In some embodiments, planning information (or a portion thereof) may be generated by processing device 140A. By way of example only, for each of at least one field, processing device 140A may obtain an initial flux map for the field and optimize the initial flux map to generate an optimal flux map for the field. The optimal flux map for each of at least one field may be considered planning information related to at least one field. In some embodiments, processing device 140A may optimize the initial flux map according to a flux map algorithm described elsewhere in this specification.

[0087] At 520 , the processing device 140A (eg, the generation module 402 ) may generate inputs to a flux map generation model based on the planning information.

[0088] As described herein, a flux map generation model refers to a model (e.g., a machine learning model) or algorithm that generates an implementation flux map based on its input. In some embodiments, the flux map generation model may be a machine learning model. For example, the flux map generation model may include a neural network model, such as a convolutional neural network (CNN) model (e.g., a full CNN model, a V-net model, a U-net model, an AlexNet model, an Oxford Visual Geometry Group (VGG) model, a ResNet model), a Generative Adversarial Network (GAN) model, etc., or any combination thereof. In some embodiments, the flux map generation model may include one or more components for feature extraction and / or feature combination, such as a full convolution block, a skip connection, a residual block, a dense block, etc., or any combination thereof.

[0089] In some embodiments, the processing device 140A may obtain the flux map generation model via a network (e.g., network 120) through one or more components of the RT planning system 100 (e.g., storage device 150, terminal 130) or an external source. For example, the flux map generation model may be pre-trained by a computer device (e.g., processing device 140B) and stored in a storage device of the RT planning system 100 (e.g., storage device 220 and / or memory 390). The processing device 140A may access the storage device and retrieve the flux map generation model from the storage device. In some embodiments, the flux map generation model may be generated according to a machine learning algorithm. The machine learning algorithm may include, but is not limited to, an artificial neural network algorithm, a deep learning algorithm, a decision tree algorithm, an association rule algorithm, an inductive logic programming algorithm, a support vector machine algorithm, a clustering algorithm, a Bayesian network algorithm, a reinforcement learning algorithm, a representation learning algorithm, a similarity and metric learning algorithm, a sparse dictionary learning algorithm, a genetic algorithm, a rule-based machine learning algorithm, etc., or any combination thereof. The machine learning algorithm used to generate one or more machine learning models may be a supervised learning algorithm, a semi-supervised learning algorithm, an unsupervised learning algorithm, etc. In some embodiments, the process for generating a flux map generation model disclosed herein (eg, process 900 ) may be executed by a computer device (eg, processing device 140B) to generate a flux map generation model.

[0090] In some embodiments, the inputs to the flux map generation model may include a reference image of the object, segmentation information of the ROI of the object, and a field angle of each field in at least one field. Alternatively, the inputs to the flux map generation model may include an optimal flux map for each field in at least one field. In some embodiments, at least one field may include at least two fields. The processing device 140A may determine the inputs for each field separately based on the planning information. By way of example only, for each field, the processing device 140A may acquire or generate a reference image, segmentation information of the ROI, and a field angle of the field as inputs for the corresponding field. Alternatively, at least two fields may share the same inputs. By way of example only, the processing device 140A may acquire or generate an optimal flux map for at least two fields as inputs for the corresponding at least two fields. More description of the inputs to the flux map generation model may be found elsewhere in this specification. For example, see Figures 6 to 8 and its related description.

[0091] At 530 , for each of the at least one field, the processing device 140A (eg, module 402 ) may generate at least one implemented flux map associated with at least two sub-fields of the field based on the inputs and the flux map generation model.

[0092] As described above, at least one implementation flux map associated with a sub-field of a field may include, for example, a composite flux map of at least two sub-fields, a sub-field flux map for each sub-field, or any combination. In some embodiments, the processing device 140A may input the input determined in 520 into a flux map generation model, and the flux map generation model may output at least one implementation flux map for each field in response to the input. For example, for one field, the flux map generation model may output a composite flux map of sub-fields of the field or at least two sub-field flux maps, each sub-field flux map corresponding to a sub-field of the field. In some embodiments, a field may include only one sub-field, and the at least one implementation flux map of the field may include a sub-field flux map of the sub-field.

[0093] Alternatively, the flux map generation model may generate an output in response to an input, and the processing device 140A may generate at least one implementation flux map for the field based on the output of the flux map generation model. For example, the output of the flux map generation model may include a composite flux map of the field, and the processing device 140A may convert the composite flux map into at least two sub-field flux maps of at least two sub-fields of the field based on a leaf sequencing algorithm.

[0094] In some embodiments, at least one field may include at least two fields. In 520, the processing device 140A may determine an input corresponding to each field. In 530, for each field, the processing device 140A may input the corresponding input into a flux map generation model to obtain at least one implemented flux map of the field, or obtain an output from the flux map generation model and generate at least one implemented flux map of the field based on the output. Alternatively, at least two fields may share the same input. The processing device 140A may input the input into the flux map generation model, and the flux map generation model may jointly output at least one implemented flux map for each field. More description of generating at least one implemented flux map for each field may be found elsewhere in this specification. For example, see Figures 6 to 8 and its related description.

[0095] It should be noted that the above description of process 500 is for illustrative purposes only and is not intended to limit the scope of the present application. For those of ordinary skill in the art, various changes and modifications can be made based on the description of the present application. However, these changes and modifications do not depart from the scope of the present application. In some embodiments, process 500 can be completed by one or more additional operations not described and / or without one or more of the above operations. For example, process 500 may include additional operations before 510 to obtain one or more flux maps associated with at least one field.

[0096] Figure 6 is a schematic diagram of an exemplary flux map generation model according to some embodiments of the present specification.

[0097] like Figure 6 As shown, the input 610 of the flux map generation model 620 may include a reference image of the object, segmentation information of one or more ROIs of the object, and a field angle of each field to be applied to the object during treatment. For illustration purposes, it is assumed that N fields may be applied to the object during treatment, where n can be any positive integer. In this case, N field angles (denoted as G1, G2, ... and GN) of the N fields may be obtained as part of the input 610.

[0098] As described in 510, the segmentation information of the ROI may include the outline of the ROI segmented from a reference image (or other image) of the object, one or more parameters describing the outline of the ROI (e.g., shape, height, width, thickness, area, ratio of height to width), etc., or any combination thereof.

[0099] In some embodiments, the outline of the ROI can be manually, semi-automatically, or automatically segmented from a reference image (or other image) of the object. In a manual method, the outline of the ROI can be segmented from the reference image (or other image) according to instructions provided by a user. For example, through a user interface implemented on, for example, terminal 130 or mobile device 300, a user can mark the outline in the reference image. In a semi-automatic method, the outline of the ROI can be segmented from the reference image (or other image) with user intervention by a computer device (e.g., Figure 2 The computer device 200 shown in FIG. 2 identifies the contour of the ROI from a reference image (or other image). For example, contour segmentation can be performed by the computer device based on an image segmentation algorithm in combination with information provided by a user. Exemplary user intervention for the semi-automatic method may include providing parameters related to the image segmentation algorithm, providing position parameters related to the ROI, adjusting or confirming the initial contour segmentation performed by the computer device, providing instructions for the computer device to repeat or redo the contour segmentation, etc. In the automatic method, the contour segmentation can be performed by the computer device (e.g., as shown in FIG. 2 ) without user intervention. Figure 2 The computer device 200 shown in FIG. 2 automatically identifies the contour of the ROI from the reference image. For example, the contour can be automatically identified from the reference image by image analysis (such as according to an image segmentation algorithm, a feature recognition algorithm, etc. or any combination thereof).

[0100] In some embodiments, the input 610 may be directly input into the flux map generation model 620. Alternatively, the input 610 may be pre-processed, and the pre-processed input may be input into the flux map generation model 620. For example, the processing device 140A may perform one or more image processing operations on the reference image to pre-process the reference image, such as image denoising, image enhancement, image smoothing, image transformation, image resampling, image normalization, etc. or any combination thereof. For example only, the processing device 140A may determine whether the imaging resolution of the reference image is the same as (or substantially the same as) a preset image resolution. If the reference image has an imaging resolution different from the preset image resolution, the processing device 140A may resample the reference image to generate a resampled reference image having the preset imaging resolution.

[0101] The output 630 of the flux map generation model 620 may include a composite flux map (denoted as I1, I2, ..., and IN) corresponding to each of the N fields. For example, the composite flux map Ii may correspond to at least two sub-fields of the i-th field with field angle Gi, where i may be any positive integer equal to or less than N. In some embodiments, the composite flux map Ii may be converted into at least two sub-field flux maps according to, for example, a leaf sequencing algorithm, each sub-field flux map corresponding to a sub-field of the i-th field.

[0102] In some embodiments, N may be greater than 1, and at least two fields will be applied to the object during treatment. The input 610 of the flux map generation model 620 may include an input corresponding to each field. For each field, the corresponding input may be input into the flux map generation model 620 to obtain a composite flux map of a sub-field of the field. For example, for the i-th field, the corresponding input may include a reference image of the object, segmentation information of the ROI of the object, and a field angle Gi of the i-th field. The input of the i-th field may be input into the flux map generation model 620, and the flux map generation model 620 may output a composite flux map Ii corresponding to the i-th field.

[0103] In some embodiments, at least two fields may share the same input 610, which includes a reference image, segmentation information of the ROI, and a field angle of each field. The input 610 may be input into a flux map generation model 620, and the flux map generation model 620 may jointly output a composite flux map of at least two fields. In some embodiments, the flux map generation model 620 may be trained to learn the interaction or relationship between at least two fields. For example, during treatment, the total radiation dose applied to the object by all fields may need to meet specific constraints, for example, below a threshold dose. The radiation doses applied by different fields may have a compensatory relationship, for example, if the radiation dose applied by a certain field is relatively high, the radiation dose applied by the remaining fields may be relatively low. In generating a composite flux map of the fields, the flux map generation model 620 may consider the compensatory relationship between the fields, which may reduce radiation damage to the object and improve treatment accuracy. In some embodiments, the flux map generation model 620 may learn the compensatory relationship from training data. Alternatively, the training data for the flux map generation model 620 may include one or more parameters related to the compensation relationship, such as a threshold dose for the total radiation dose applied by the field. Optionally, the threshold dose may be part of the input 610 when applying the flux map generation model 620.

[0104] In some embodiments, at least two flux map generation models 620 may be applied, each corresponding to a specific part of a human being (e.g., a specific organ). For example, the processing device 140A may select a flux map generation model 620 from the at least two flux map generation models 620 based on the target area of ​​the object to be treated. The selected flux map generation model 620 may be used for the treatment plan of the object. By way of example only, there may be a first flux map generation model 620 corresponding to the heart, a second flux map generation model 620 corresponding to the abdomen, and a third flux map model 620 corresponding to the head. If the target area is in the head of the object, the processing device 140A may select the third flux map generation model 620 for generating an implementation flux map for the object. As described elsewhere in this specification (e.g., Figure 5As described in the related descriptions), conventional treatment planning techniques may require generating an initial flux map of a field, optimizing the initial flux map to generate an optimal flux map of the field, and further dividing the optimal flux map of the field into implementation flux maps. The division of the optimal flux map of the field usually involves an iterative or manual optimization process of at least two initial sub-field flux maps. According to some embodiments of the present specification, a reference image can be combined with other information and input into a flux map generation model 620 to generate at least one implementation flux map of the field. The application of the flux map generation model 620 can avoid the operations of generating an initial flux map of the field, optimizing the initial flux map, and splitting the optimal flux map, which can improve the efficiency of the treatment plan by reducing the workload of the user, the impact of user differences, and / or the time required for the treatment plan. In addition, in some embodiments, a specific flux map generation model 620 can be selected according to the target area of ​​the object and used for the treatment plan. Compared with using the same flux map generation model for different target areas, the system and method disclosed herein can improve the accuracy of at least one implementation flux map generated, thereby improving the accuracy of radiotherapy.

[0105] Figure 7 is a schematic diagram of an exemplary flux map generation model according to some embodiments of the present specification.

[0106] like Figure 7 As shown, the input 710 of the flux map generation model 720 may include an optimal flux map for each of at least one field to be applied to the object. For example, N fields may be applied to the object, and N optimal flux maps (denoted as F1, F2, ... and FN) may be specified as input 710. In some embodiments, for each field, an optimal flux map of the field may be generated by optimizing an initial flux map of the field by the processing device 140A. For example, an initial flux map of the field may be generated based on a planning image (e.g., a CT image) of the object, and the processing device 140A may optimize the initial flux map based on a flux map optimization algorithm to generate an optimal flux map for the field. More description on generating optimal flux maps may be found elsewhere in this specification. For example, see Figure 5 and related operations thereof. In some embodiments, the optimal flux map may be pre-generated and stored in a storage device (e.g., the storage device 150 or an external storage device). The processing device 140A may obtain the optimal flux map of the field from the storage device and specify the obtained optimal flux map as the input 710 or a part of the input 710.

[0107] In some embodiments, the input 70 may be directly input into the flux map generation model 720. Alternatively, the input 710 may be pre-processed (eg, resampled), and the pre-processed input may be input into the flux map generation model 720.

[0108] The output 730 of the flux map generation model 720 may include composite flux maps I1, I2, ..., and IN, and each composite flux map corresponds to one of the N fields. The output 730 may be similar to Figure 6 The output 630 is not described again here.

[0109] In some embodiments, N may be greater than 1, and at least two fields may be planned to be applied to the subject during treatment. The input 710 of the flux map generation model 720 may include inputs corresponding to each field. For example, the input of the i-th field may include an optimal flux map Fi. The input of the i-th field may be input into the flux map generation model 720, and the flux map generation model 720 may output a composite flux map Li corresponding to the i-th field. Alternatively, at least two fields may share the same input 710, which may include an optimal flux map for each field. The optimal flux maps of the fields may be simultaneously input into the flux map generation model 720, and the flux map generation model 720 may jointly output a composite flux map of the fields.

[0110] Figure 8 is a schematic diagram of an exemplary flux map generation model according to some embodiments of the present specification.

[0111] like Figure 8 As shown, the input 810 of the flux map generation model 820 can be similar to the input 710 of the flux map generation model 720, such as Figure 7 The input 810 may include the optimal flux maps F1, F2, ... and FN for the N fields.

[0112] The output 830 of the flux map generation model 820 may include at least two sub-field flux maps for each field. The at least two sub-field flux maps of a field may correspond to at least two sub-fields of the field. For example, the i-th field may have Mi sub-fields, where Mi may be any positive integer greater than 1. The number of sub-fields of different fields may be the same or different. Figure 8 As shown, the output 830 of the flux map generation model 820 may include a sub-field flux map I corresponding to the i-th field. i-1 , I i-2 , ..., I i-Mi . Each sub-field flux map I i-1 , I i-2 , ..., I i-Mi It can correspond to a sub-field in the Mi sub-field of the i-th field.

[0113] In some embodiments, the optimal flux map of each field can be individually input into the flux map generation model 820, and the flux map generation model 820 can output the sub-field flux maps of different fields respectively. Alternatively, the optimal flux maps of the fields can be simultaneously input into the flux map generation model 820, and the flux map generation model 820 can jointly output the sub-field flux maps of different fields.

[0114] As described elsewhere in this specification (e.g. Figure 5 and related descriptions), conventional treatment planning techniques may require generating an initial flux map of a field, optimizing the initial flux map to generate an optimal flux map of the field, and further dividing the optimal flux map of the field into implementation flux maps. The division of the optimal flux map of the field generally involves an iterative or manual optimization process of at least two initial sub-field flux maps. Compared to conventional treatment planning techniques, the systems and methods utilizing the flux map generation model 720 or 820 disclosed herein do not require iterative or manual updating of the initial sub-field flux map, which can improve the efficiency of treatment planning, for example, reducing the user's workload, cross-user conversions, and / or the time required for treatment planning.

[0115] Combination Figure 7 As described above, the flux map generation model 720 can output a composite flux map of a field, and may need to convert the composite flux map into at least two sub-field flux maps corresponding to at least two sub-fields of the field. Compared with the flux map generation model 720, the flux map generation model 820 can directly output the sub-field flux map of the field without converting the composite flux map of the field, and improve the efficiency of the treatment plan. In some embodiments, combined with Figure 6 As described above, at least two flux map generation models 620 corresponding to different human body parts can be trained, and a specific flux map generation model 620 can be selected according to the target area of ​​the object. Compared with the flux map generation model 620, the flux map generation model 720 and / or the flux map generation model 820 can have higher universality and be suitable for different human body parts.

[0116] It should be noted that Figures 6 to 8The above examples shown in are provided for illustrative purposes only and are not intended to limit the scope of this specification. For those of ordinary skill in the art, various changes and modifications can be made according to the description of this specification. However, these changes and modifications do not depart from the scope of this specification. In some embodiments, the input of the flux map generation model may include additional information, such as dose constraints, prescriptions for objects specified by users (e.g., doctors, radiologists). Additionally or alternatively, a portion of the input as described above may be omitted. As an example only, the reference image in the input 610 of the flux map generation model 620 may be omitted. In some embodiments, the output of the flux map generation model may include additional information or omit specific information as described above. For example, the output 630 of the flux map generation model 620 and / or the output 730 of the flux map generation model 720 may include at least two sub-field flux maps for each field. As another example, the output 830 may include a composite flux map for each field.

[0117] Fig. 9 4b ) can execute the set of instructions and thus be directed to execute process 900. In some embodiments, one or more operations of process 900 can be performed to implement the following: Figure 5 At least a portion of step 520 described above. In some embodiments, process 900 may be performed by another device or system other than RT planning system 100, such as a device or system of a manufacturer or supplier. For illustrative purposes, the implementation process of process 900 is described below using processing device 140B as an example.

[0118] In 910 , the processing device 140B (eg, the acquisition module 403 ) may acquire at least one training sample.

[0119] In some embodiments, the training sample may include sample plan information and at least one implementation flux map truth. The sample plan information of the training sample may involve at least one sample field to be applied to the sample object. The sample object may be of the same type or a different type than the object described in 510. For example, the object may be a patient's head, and the sample object may be another patient's head (or another part) or an artificial object (e.g., a phantom). The sample plan information of the sample object refers to the plan information of the sample object. For example, the sample plan information may include sample feature information of the sample object, sample feature information of one or more sample ROIs of the sample object (e.g., sample segmentation information), sample field angles of each sample field in at least one sample field, a sample reference image of the sample object, an optimal sample flux map of each of at least one sample field, a dose constraint, a prescription prescribed by a user (e.g., a doctor, a radiologist) to the sample object, or the like, or any combination thereof. The sample segmentation information of the sample ROI refers to the segmentation information of the sample ROI. The sample field angle of the sample field refers to the field angle of the sample field. The sample reference image of the sample object refers to a reference image (e.g., a planning image) of the sample object.

[0120] At least one implementation flux map truth value of a training sample may relate to at least one sample sub-field of at least one sample field. For example, the sample field of a training sample may include at least two sample sub-fields, and at least one implementation flux map truth value associated with at least two sample sub-fields of the sample field may include a truth composite flux map of the sample sub-field, a truth sub-field flux map of each sample sub-field, or any combination thereof. As another example, the sample field of a training sample may include only one sample sub-field, and at least one implementation flux map truth value of the training sample may include a truth sub-field flux map of the sample sub-field. For illustrative purposes, the sample field of a training sample described below includes at least two sample sub-fields, which is not intended to limit the scope of this specification.

[0121] In some embodiments, the training sample (or a portion thereof) may be generated by a computer device (e.g., processing device 140B) and stored in a storage device (e.g., storage device 150, storage device 220, memory 390, or an external database). Processing device 140B may retrieve the training sample (or a portion thereof) from the storage device. Alternatively, the training sample (or a portion thereof) may be generated by processing device 140B. For example only, processing device 140B may perform the following steps: Fig.10 One or more steps of process 1000 are described to generate at least one ground truth flux map for training samples.

[0122] In some embodiments, a first flux map generation model that is the same as or similar to the flux map generation model 620 may be generated by executing process 900. In this case, the sample plan information of the training sample may include a sample reference image of the sample object, sample segmentation information of the sample ROI of the sample object, a sample field angle of each sample field applied to the sample object, etc., or any combination thereof. At least one implementation flux map truth value of the training sample may include a truth composite flux map of each sample field.

[0123] In some embodiments, a second flux map generation model that is the same as or similar to the flux map generation model 720 can be generated by executing process 900. In this case, the sample plan information of the training sample may include an optimal sample flux map for each sample field applied to the sample object. At least one implementation flux map truth value of the training sample may include a truth composite flux map for each sample field.

[0124] In some embodiments, a third flux map generation model that is the same as or similar to the flux map generation model 820 may be generated by executing process 900. In this case, the sample plan information of the training sample may include an optimal sample flux map for each sample field applied to the sample object. The at least one implementation flux map truth value of the training sample may include at least two truth sub-field flux maps for each sample field.

[0125] In some embodiments, for different training samples of the flux map generation model (e.g., the first, second, and third flux map generation models), the target areas of the corresponding sample objects may be the same or substantially the same. For example, for the first flux map generation model, the target areas of the sample objects of different training samples may be the same or substantially the same. By way of example only, a plurality of training samples of a plurality of sample patients with lung cancer may be used to generate a first flux map generation model corresponding to a human lung. The first flux map generation model corresponding to a human lung may be used to generate a treatment plan for treating a lung cancer patient. As yet another example, for the second or third flux map generation model, the target areas of the sample objects of different training samples may be the same or different.

[0126] In 920 , the processing device 140B (eg, the training module 404 ) may generate a flux map generation model by training an initial model using at least one training sample.

[0127] In some embodiments, the initial model can be any type of model (e.g., a machine learning model), for example, a neural network model (e.g., a CNN model, a GAN model), etc. The initial model may include one or more model parameters. For example, the initial model may be a CNN model, and exemplary model parameters of the initial model may include the number of layers (or counts), the number of kernels (or counts), the kernel size, the stride, the padding of each convolutional layer, the loss function, etc., or any combination thereof. Before training, the model parameters of the initial model may have respective initial values. For example, the processing device 140B may initialize the parameter values ​​of the model parameters of the initial model.

[0128] In some embodiments, the initial model can be based on the machine learning algorithms described elsewhere in the present invention (e.g., Figure 5 and related descriptions). For example, the processing device 140B may iteratively update the model parameters of the initial model by performing one or more iterations according to the supervised machine learning algorithm, thereby generating a flux map generation model. For illustrative purposes, an exemplary current iteration of iterations is described in the following description. The current iteration may be performed based on at least a portion of the plurality of training samples. In some embodiments, the same or different sets of training samples may be used in different iterations in training the initial model.

[0129] In the current iteration, for each of at least a portion of the training samples, the processing device 140B may generate or obtain a sample input based on the sample plan information of the training sample. The generation method of the sample input of the training sample may be similar to the generation method of the input of the flux map generation model described in step 520, and will not be repeated here. For each of at least a portion of the training samples, the processing device 140B may generate at least one predicted implementation flux map by inputting the sample input of the training sample into the updated initial model determined in the previous iteration. Then, the processing device 140B may determine the value of the loss function of the updated initial model based on at least one predicted implementation flux map and at least one implementation flux map true value of each of at least a portion of the training samples. The loss function can be used to evaluate the accuracy and reliability of the updated initial model. For example, the smaller the loss function, the more reliable the updated initial model. Exemplary loss functions may include L1 loss function, focal loss function, logarithmic loss function, cross entropy loss function, Dice loss function, etc. The processing device 140B may also update the values ​​of the model parameters of the updated initial model based on the value of the loss function according to an algorithm such as back propagation for the next iteration.

[0130] In some embodiments, if a termination condition is met in the current iteration, one or more iterations may be terminated. An exemplary termination condition may be that the value of the loss function obtained in the current iteration is less than a predetermined threshold. Other exemplary termination conditions may include a count of iterations having been executed a certain number of times, convergence of the loss function, a difference in the values ​​of the loss function obtained in consecutive iterations being within a threshold, etc. If the termination condition is met in the current iteration, the processing device 140B may designate the updated initial model as the flux map generation model.

[0131] It should be noted that the above description of process 900 is for illustrative purposes only and is not intended to limit the scope of this specification. For those of ordinary skill in the art, various changes and modifications may be made according to the description of this application. However, these changes and modifications do not depart from the scope of this application. In some embodiments, process 900 may be completed by one or more additional operations not described and / or one or more of the operations not discussed above. For example, after generating the flux map generation model, the processing device 140B may further test the flux map generation model using a set of test samples. Additionally or alternatively, the processing device 140B may periodically or irregularly update the flux map generation model based on one or more newly generated training samples (e.g., a new treatment plan generated during treatment). As yet another example, the training samples (or portions thereof) may be preprocessed before the training of the initial model. By way of example only, one or more image processing operations (e.g., image cropping, image resampling) may be performed on the optimal sample flux map of the sample reference image and / or the training sample.

[0132] Fig.10 4b ) can execute the set of instructions and thus be instructed to execute process 1000. In some embodiments, one or more operations of process 1000 can be performed in combination with process 100 to implement at least one implementation flux map truth value for generating training samples. In some embodiments, process 1000 can be performed by RT planning system 100. For example, process 1000 can be implemented by an instruction set (e.g., an application) stored in a storage device (e.g., storage device 150, storage device 220, and / or storage device 390). In some embodiments, processing device 140 (e.g., processor 210 of computer device 200, CPU 340 of mobile device 300, and / or one or more modules shown in FIG. 4b ) can execute the set of instructions and thus be instructed to execute process 1000. In some embodiments, one or more operations of process 1000 can be performed to implement as described in combination with process 1000. Fig. 9 At least a portion of step 910 described above.

[0133] As described in conjunction with step 910, the training sample may include sample plan information related to one or at least two sample fields applied to the sample object, and at least one implementation flux map truth associated with at least two sample sub-fields of each sample field. The at least one implementation flux map truth of the sample field may include a true composite flux map of the sample sub-field of the sample field, a true sub-field flux map of each sample sub-field of the sample field, and the like. In some embodiments, the processing device 140B may execute process 1000 for each sample field of the training sample to generate at least one implementation flux map of the sample field. For illustrative purposes, the following describes how to execute process 1000 for one sample field as an example.

[0134] At 1010 , for a sample field to be applied to a sample object, the processing device 140B (eg, such as module 405 ) may acquire an initial sample flux map of the sample field.

[0135] In some embodiments, an initial sample flux map may be generated based on a planning image of a sample object. The initial sample flux map of a sample field may be generated based on a planning image of a sample object. Figure 5 The initial flux diagram of the field is similar.

[0136] At 1020, the processing device 140B (e.g., module 405) may generate an optimal sample flux map by optimizing the initial sample flux map. For example, the optimal sample flux map may be generated based on other places in this specification (e.g., Figure 5 and related descriptions) to perform optimization of the initial sample flux map.

[0137] In 1030, the processing device 140B (e.g., such as module 405) can convert the optimal sample flux map into at least two initial sample sub-field flux maps of at least two sample sub-fields of the sample field. Each initial sample sub-field flux map can correspond to one of the sample sub-fields of the sample field. For example, the optimal sample flux map can be converted into the initial sample sub-field flux map according to the leaf sequencing algorithm.

[0138] In 1040 , the processing device 140B (eg, module 405 ) may generate at least one implemented flux map truth value corresponding to at least two sample sub-fields of the sample field based on the at least two initial sample sub-field flux maps.

[0139] In some embodiments, there may be an error between the optimal sample flux map and the composite flux map of the initial sample sub-field flux map. It may be necessary to optimize the initial sample sub-field flux map to generate at least two optimal sample sub-field flux maps so that the composite flux map of the optimal sample sub-field flux map matches the optimal sample flux map. As used herein, if the similarity between two flux maps exceeds a threshold similarity, then the two flux maps can be considered to match each other.

[0140] In some embodiments, the optimization of the initial sample sub-field flux map can be performed automatically, semi-automatically, or manually. For example, the processing device 140B can iteratively update the initial sample sub-field flux map according to one or more flux map optimization algorithms. As another example, a user (e.g., a doctor, a radiologist) can manually adjust the parameters of the sample sub-field and / or the initial sample sub-field flux map through an interface implemented on a terminal device, for example. As yet another example, the processing device 140B can iteratively update the sample sub-field flux map according to one or more flux map optimization algorithms in combination with user intervention (e.g., adjusting or confirming the preliminary optimization results generated by the processing device 140B).

[0141] After optimizing the initial sub-field flux map, the processing device 140B may designate the optimal sample sub-field flux map as the true sub-field flux map of the sample sub-field. Additionally or alternatively, the processing device 140B may designate a composite flux map of the optimal sample sub-field flux map as the true composite flux map of the sample sub-field.

[0142] It should be noted that the above description about flow 1000 is provided only for the purpose of illustration, and is not intended to limit the scope of the present application. For those of ordinary skill in the art, various changes and modifications can be made according to the description of the present application. However, these changes and modifications do not depart from the scope of the present application. In certain embodiments, flow 1000 can utilize one or more undescribed additional operations and / or do not have one or more operations discussed above to complete. For example, steps 1010 and 1020 can be integrated into one step.

[0143] The basic concepts have been described above. Obviously, for those of ordinary skill in the art who have read this application, the above invention disclosure is only for example and does not constitute a limitation of this application. Although not explicitly stated here, those of ordinary skill in the art may make various modifications, improvements and amendments to this application. Such modifications, improvements and amendments are suggested in this application, so such modifications, improvements and amendments still belong to the spirit and scope of the exemplary embodiments of this application.

[0144] At the same time, the present application uses specific words to describe the embodiments of the present application. For example, "one embodiment", "an embodiment", and "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present application. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present application can be appropriately combined.

[0145] In addition, it will be appreciated by those of ordinary skill in the art that various aspects of the present application may be illustrated and described by a number of patentable categories or situations, including any new and useful process, machine, product, or combination of substances, or any new and useful improvements thereto. Therefore, various aspects of the present application may be implemented entirely in hardware, entirely in software (including firmware, resident software, microcode, etc.), or in a combination of software and hardware implementations, all of which may be collectively referred to herein as "modules," "units," "components," "devices," or "systems." In addition, aspects of the present application may take the form of a computer program product embodied in one or more computer-readable media having computer-readable program code embodied thereon.

[0146] A computer readable signal medium may include a propagated data signal containing computer program code, for example, in baseband or as part of a carrier wave. Such propagated signals may be in a variety of forms, including electromagnetic, optical, etc., or any suitable combination. A computer readable signal medium may be any computer readable medium other than a computer readable storage medium, which may be connected to an instruction execution system, device or apparatus to communicate, propagate or transmit a program for use. The program code on a computer readable signal medium may be propagated via any suitable medium, including radio, cable, fiber optic cable, RF, etc., or any combination of the above.

[0147] Computer program code for performing the operations of various aspects of the present application may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C, C++. The program code may be run entirely on a user's computer, or as an independent software package on a user's computer, or partially on a user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., by using a network service provider's network) or in a cloud computing environment or provided as a service, e.g., software as a service (SaaS).

[0148] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this application, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this application. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this application. For example, although the implementation of the various components described above can be embodied in a hardware device, it can also be implemented as a pure software solution, for example, installation on an existing server or mobile device.

[0149] Similarly, it should be noted that in order to simplify the description disclosed in this application and thus help understand one or more embodiments of the invention, in the above description of the embodiments of the present application, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this method of the present application should not be interpreted as reflecting the intention that the claimed object material to be scanned requires more features than those explicitly stated in each claim. In fact, the features of the embodiment are less than all the features of the single embodiment disclosed above.

[0150] In some embodiments, the numbers representing the quantity or property used to describe and claim certain embodiments of the present application should be understood to be modified by the terms "approximately", "approximately" or "substantially" in some cases. For example, unless otherwise specified, "approximately", "approximately" or "substantially" may indicate certain changes in the values ​​described therein (e.g., ±1%, ±5%, ±10% or ±20%). Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the characteristics required by the individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the method of general digit retention. Although the numerical domains and parameters used to confirm the breadth of the range in some embodiments of the present application are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range. In some embodiments, classification conditions for classification are provided for illustrative purposes and are modified according to different situations. For example, the classification condition of "probability value is greater than a threshold value" may further include or exclude the condition of "probability value equal to a threshold value".

[0151] Each patent, patent application, patent application, patent application, material such as articles, books, specifications, publications, documents, things and / or the like referenced herein is incorporated by reference for all purposes, except for any prosecution document history related to this document, except to the extent that any prosecution document history is inconsistent with this document, or for any purpose that might have a limiting effect on the broadest scope of the claims present in this document. For example, if there is any inconsistency or conflict between the description, definition and / or use of terminology associated with any incorporated material and the terminology associated with this document, the description, definition and / or use of terminology in this document shall control.

[0152] Finally, it should be understood that the embodiments described in this application are only used to illustrate the principles of the embodiments of the present application. Other variations may also fall within the scope of the present application. Therefore, as an example and not a limitation, the alternative configurations of the embodiments of the present application may be considered to be consistent with the teachings of the present application. Accordingly, the embodiments of the present application are not limited to the embodiments explicitly introduced and described in the present application.

Claims

1. A radiation therapy planning system, comprising: at least one storage device including a set of instructions; as well as At least one processor is configured to communicate with the at least one storage device, and when executing the instruction set, the at least one processor is configured to instruct the system to perform the following operations: acquiring planning information related to at least one field to be applied to the subject in treatment; generating an input of a flux map generation model based on the planning information, the input comprising segmentation information of one or more regions of interest (ROI) of the object to which the field is to be applied and a field angle of each of the at least one field, or the input comprising an optimal flux map of each of the at least one field; as well as For each of the at least one field, at least one implementation flux map associated with at least two sub-fields of the field is generated based on the input and the flux map generation model, wherein each field corresponds to a field angle, and each field is applied to the object through the at least two sub-fields having different shapes at the corresponding field angle, and the flux map generation model is a trained machine learning model.

2. The system according to claim 1, characterized in that The plan information includes: Segmentation information of one or more regions of interest (ROI) of the object to which the field is to be applied and a field angle of each of the at least one field; and / or An optimal flux map for each of the at least one field.

3. The system according to claim 1 or 2, characterized in that: The planning information also includes a reference image of the object.

4. The system according to claim 1, characterized in that Each of the fields corresponds to a field angle, and each of the fields is applied to the object through the at least two sub-fields with different shapes at the corresponding field angle.

5. The system according to claim 2, characterized in that The acquisition plan information includes: For each of the at least one field, obtaining an initial flux map of the field; and An optimal flux map of the field is generated by optimizing the initial flux map.

6. The system according to claim 1, characterized in that For each of the at least one field, The at least one implemented flux map associated with at least two sub-fields of the field comprises a composite flux map of the at least two sub-fields.

7. The system according to claim 6, characterized in that The at least one processor is further configured to instruct the system to perform the following operations, including: For each of the at least one field, the composite flux map is converted into at least two sub-field flux maps, each of the sub-field flux maps corresponding to one of the at least two sub-fields.

8. The system according to claim 1, characterized in that For each of the at least one field, the at least two implemented flux maps associated with at least two sub-fields of the field include at least two sub-field flux maps, each of the sub-field flux maps corresponding to one of the at least two sub-fields.

9. The system according to claim 1, characterized in that The flux map generation model includes at least one of a convolutional neural network (CNN) or a generative adversarial network (GAN).

10. The system according to claim 1, characterized in that The flux map generation model is generated through the following training process: Acquire at least one training sample, wherein each training sample includes sample plan information and at least one implementation flux map true value, the sample plan information is related to at least one sample field to be applied to the sample object, and the at least one implementation flux map true value is related to at least two sample sub-fields of the at least one sample field; as well as The flux map generation model is generated by training an initial model using the at least one training sample.

11. The system according to claim 10, characterized in that For each of the at least one training sample, obtaining the training sample comprises: For each of the at least one training sample corresponding to the at least one sample field, Obtaining an initial sample flux map of the sample field; Generating an optimal sample flux map of the sample field by optimizing the initial sample flux map; converting the optimal sample flux map into at least two initial sample sub-field flux maps of at least two sample sub-fields of the sample field; and Based on the at least two initial sample sub-field flux maps, the at least one implemented flux map truth value associated with the at least two sample sub-fields is generated.

12. A method implemented on a computer device comprising at least one processor and at least one computer readable storage medium for radiation therapy planning, the method comprising: acquiring planning information related to at least one field to be applied to the subject in treatment; generating an input of a flux map generation model based on the planning information, the input comprising segmentation information of one or more regions of interest (ROI) of the object to which the field is to be applied and a field angle of each of the at least one field, or the input comprising an optimal flux map of each of the at least one field; as well as For each of the at least one field, at least one implementation flux map related to at least two sub-fields of the field is generated based on the input and the flux map generation model, wherein each field corresponds to a field angle, and each field is applied to the object through the at least two sub-fields with different shapes at the corresponding field angle, and the flux map generation model is a trained machine learning model.

13. The method according to claim 12, characterized in that The plan information includes: Segmentation information of one or more regions of interest (ROI) of the object to which the field is to be applied and a field angle of each of the at least one field; and / or An optimal flux map for each of the at least one field.

14. The method according to claim 12 or 13, characterized in that The planning information also includes a reference image of the object.

15. The method according to claim 12, characterized in that Each of the fields corresponds to a field angle, and each of the fields is applied to the object through the at least two sub-fields with different shapes at the corresponding field angle.

16. The method according to claim 13, characterized in that The acquisition plan information includes: For each of the at least one field, obtaining an initial flux map of the field; and An optimal flux map of the field is generated by optimizing the initial flux map.

17. The method according to claim 12, characterized in that For each of the at least one field, The at least one implemented flux map associated with at least two sub-fields of the field comprises a composite flux map of the at least two sub-fields.

18. The method of claim 17, further comprising: For each of the at least one field, the composite flux map is converted into at least two sub-field flux maps, each of the sub-field flux maps corresponding to one of the at least two sub-fields.

19. The method according to claim 12, characterized in that For each of the at least one field, the at least one implemented flux map associated with the at least two sub-fields of the field includes at least two sub-field flux maps, each of the sub-field flux maps corresponding to one of the at least two sub-fields.

20. The method according to claim 12, characterized in that The flux map generation model includes at least one of a convolutional neural network (CNN) or a generative adversarial network (GAN).

21. The method according to claim 12, characterized in that The flux map generation model is generated through the following training process: Acquire at least one training sample, wherein each training sample comprises sample plan information and at least one implementation flux map true value, wherein the sample plan information is associated with at least one sample field to be applied to the sample object, and the at least one implementation flux map true value is associated with at least two sample sub-fields of the at least one sample field; and The flux map generation model is generated by training an initial model using the at least one training sample.

22. The method according to claim 21, characterized in that For each of the at least one training sample, obtaining the training sample comprises: For each of the at least one training sample corresponding to the at least one sample field, Obtaining an initial sample flux map of the sample field; Generating an optimal sample flux map of the sample field by optimizing the initial sample flux map; converting the optimal sample flux map into at least two initial sample sub-field flux maps of at least two sample sub-fields of the sample field; and Based on the at least two initial sample sub-field flux maps, the at least one implemented flux map truth value associated with the at least two sample sub-fields is generated.

23. A non-transitory computer readable storage medium comprising a set of instructions for radiation therapy planning, which, when executed by at least one processor, directs the at least one processor to implement the following method, comprising: acquiring planning information related to at least one field to be applied to the subject in treatment; generating an input of a flux map generation model based on the planning information, the input comprising segmentation information of one or more regions of interest (ROI) of the object to which the field is to be applied and a field angle of each of the at least one field, or the input comprising an optimal flux map of each of the at least one field; as well as For each of the at least one field, at least one implementation flux map related to at least two sub-fields of the field is generated based on the input and the flux map generation model, wherein each field corresponds to a field angle, and each field is applied to the object through the at least two sub-fields with different shapes at the corresponding field angle, and the flux map generation model is a trained machine learning model.

Images