Quality assurance method and system

Abstract

The embodiment of the invention provides a quality assurance method and system. The method comprises the following steps: acquiring state information of medical equipment; obtaining a target plan of the target object; determining a prediction result based on the state information of the medical equipment and the target plan; and determining whether the quality assurance test passes based on the prediction result.

Description

Quality assurance methods and systems

[0001] Cross-references

[0002] This application claims priority to Chinese application No. 202210611450.9 filed on May 31, 2022, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure relates generally to radiation therapy systems and methods, and more particularly to systems and methods for quality assurance in radiation therapy. Background Art

[0004] Radiation therapy is a crucial treatment modality in cancer treatment. To ensure accurate execution of radiotherapy plans, quality assurance testing of the patient's radiotherapy plan is essential. However, the quality assurance testing workflow for radiotherapy plans is time-consuming, third-party quality assurance testing equipment is often expensive and cumbersome to use, and when a radiotherapy plan fails, analyzing the reasons for the failure is difficult. Therefore, it is desirable to provide a quality assurance system and method to improve the efficiency of quality assurance testing.

[0005] Summary of the Invention

[0006] According to one aspect of the present specification, a quality assurance method is provided, comprising: obtaining status information of a medical device; obtaining a target plan for a target object; determining a prediction result based on the status information of the medical device and the target plan; and determining whether a quality assurance test passes based on the prediction result.

[0007] In some embodiments, the status information of the medical device includes at least one of beam information corresponding to multiple dose rates, positioning accuracy information of at least one component of the medical device at multiple positions, or operating error of at least one component of the medical device.

[0008] In some embodiments, obtaining status information of a medical device includes: obtaining at least one data set, wherein the at least one data set is determined based on at least one of a candidate plan, a limit performance of at least one component of the medical device, or a calculation limit of a dose model; and determining the status information of the medical device by executing the at least one data set using the medical device.

[0009] In some embodiments, determining the prediction result based on the status information of the medical device and the target plan includes: determining whether the status information of the medical device meets a preset condition; and in response to determining that the status information of the medical device meets the preset condition, determining the prediction result based on the status information of the medical device and the target plan.

[0010] In some embodiments, the at least one data set includes at least one of a first data set, a second data set, or a third data set, and acquiring the at least one data set includes: determining the first data set based on at least one first candidate plan, wherein the value of the first candidate parameter in the at least one first candidate plan is within a first preset range; determining the second data set based on at least one second candidate plan, wherein the value of the second candidate parameter in the at least one second candidate plan is outside a second preset range; and determining the third data set based on the limit performance of at least one component of the medical device and / or the computational limit of the dose model.

[0011] In some embodiments, the at least one data set is the second data set or the third data set, and determining whether the status information of the medical device meets the preset conditions includes: determining multiple test results by performing multiple tests based on the at least one data set; and determining whether the status information of the medical device meets the preset conditions based on the multiple test results.

[0012] In some embodiments, obtaining a target plan of the target object includes: determining, based on the target plan, features and / or a target flux map related to a complexity of the target plan.

[0013] In some embodiments, determining the prediction result based on the status information of the medical device and the target plan includes: determining the prediction result according to a quality assurance model based on the status information of the medical device and the target plan of the target object, wherein the quality assurance model is a machine learning model.

[0014] In some embodiments, the quality assurance model includes a first model, and the prediction result is determined according to the quality assurance model based on the status information of the medical device and the target plan of the target object, including: determining the prediction result by inputting the status information of the medical device and the target plan of the target object into the first model, wherein the prediction result includes at least one of a predicted image, a gamma pass rate or a dose distribution.

[0015] In some embodiments, the quality assurance model includes a second model, and the method further includes: in response to determining that the quality assurance test fails, determining the reason why the quality assurance test fails according to the second model based on the status information of the medical device, the target plan of the target object and the prediction result.

[0016] In some embodiments, the method further includes: adjusting corresponding parameter values ​​based on the reason why the quality assurance test failed; determining an updated prediction result based on the first model based on the adjusted parameter values; determining whether the quality assurance test passed based on the updated prediction result; and in response to determining that the quality assurance test passed, controlling the medical device to execute an updated target plan on the target object, wherein the updated target plan is determined based on the adjusted parameter values.

[0017] In some embodiments, the target object includes multiple regions of interest, the prediction results include dose distributions corresponding to the multiple regions of interest, and determining whether the quality assurance test is passed based on the prediction results includes: obtaining a weight value corresponding to each region of interest in the multiple regions of interest; and determining whether the quality assurance test is passed based on the weight value corresponding to each region of interest in the multiple regions of interest and the dose distributions corresponding to the multiple regions of interest.

[0018] In some embodiments, determining the prediction result based on the status information of the medical device and the target plan includes: acquiring an image of the target object; and determining the three-dimensional dose distribution in the target object according to a quality assurance model based on the status information of the medical device, the target plan of the target object and the image of the target object.

[0019] According to another aspect of the present disclosure, a quality assurance system is provided. The system includes: at least one storage device storing an instruction set; and at least one processor in communication with the at least one storage device. When executing the instruction set, the at least one processor is configured to cause the system to perform the following operations: obtain status information of a medical device; obtain a target plan for a target subject; determine a prediction result based on the status information of the medical device and the target plan; and determine whether a quality assurance test passes based on the prediction result.

[0020] According to another aspect of the present disclosure, a non-transitory computer-readable medium is provided. The non-transitory computer-readable medium includes at least one set of instructions that, when executed by at least one processor of a computing device, cause the at least one processor to implement a method comprising: obtaining status information of a medical device; obtaining a target plan for a target subject; determining a prediction result based on the status information of the medical device and the target plan; and determining whether a quality assurance test passes based on the prediction result.

[0021] According to another aspect of the present disclosure, a quality assurance system is provided. The system includes: a first acquisition module configured to acquire status information of a medical device; a second acquisition module configured to acquire a target plan for a target subject; a determination module configured to determine a prediction result based on the status information of the medical device and the target plan; and a determination module configured to determine whether a quality assurance test has passed based on the prediction result.

[0022] According to another aspect of the present specification, a quality assurance device is provided, wherein the quality assurance device is configured to execute the above-mentioned quality assurance method.

[0023] Some additional features of this specification may be explained in the following description. Some additional features of this specification will be apparent to those skilled in the art through study of the following description and accompanying drawings, or through understanding the production or operation of the embodiments. The features of this specification may be realized and achieved through practice or use of the methods, means, and combinations of various aspects of the specific embodiments described below. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] This specification will be further described by exemplary embodiments. These exemplary embodiments will be described in detail by the accompanying drawings. These embodiments are non-limiting exemplary embodiments, and in these embodiments, the same numbers in the various figures represent similar structures, wherein:

[0025] FIG1 is a schematic diagram of an exemplary medical system according to some embodiments of the present specification;

[0026] 2 is a schematic diagram of an exemplary computing device on which a processing device may be implemented according to some embodiments of the present specification;

[0027] FIG3 is a diagram illustrating exemplary hardware and / or software components of an exemplary mobile device on which a terminal may be implemented according to some embodiments of the present specification;

[0028] FIG4 is a block diagram of an exemplary processing device according to some embodiments of the present specification;

[0029] FIG5 is a flowchart of an exemplary process of a quality assurance method according to some embodiments of the present specification;

[0030] FIG6 is a flowchart of an exemplary process of a quality assurance method according to some embodiments of the present specification; and

[0031] FIG7 is a flowchart of an exemplary process of joint training of a first model and a second model according to some embodiments of the present specification. DETAILED DESCRIPTION

[0032] In order to more clearly illustrate the technical solutions of the embodiments of this specification, a brief introduction to the drawings required for use in the description of the embodiments will be given below. However, it should be understood by those skilled in the art that this specification can be implemented without these details. In other cases, in order to avoid unnecessarily obscuring various aspects of this specification, well-known methods, processes, systems, components and / or circuits have been described at a higher level. It is obvious to those skilled in the art that various changes can be made to the disclosed embodiments, and the general principles defined in this specification can be applied to other embodiments and application scenarios without departing from the principles and scope of this specification. Therefore, this specification is not limited to the embodiments shown, but is in the broadest scope consistent with the scope of the patent application.

[0033] The terms used in this specification are for the purpose of describing specific example embodiments only and are not intended to be limiting. As used in this specification, the singular forms "a," "an," and "the" may also include the plural forms unless the context clearly indicates an exception. It should also be understood that the terms "comprises" and "includes" as used in this specification merely indicate the presence of the described features, steps, operations, components, and / or parts, but do not exclude the presence or addition of other features, steps, operations, components, parts, and / or combinations thereof.

[0034] It is understood that the terms "system," "engine," "unit," "module," and / or "block" used in this specification are methods for distinguishing different components, elements, parts, portions, or assemblies at different levels in ascending order. However, these terms may be replaced by other expressions if the same purpose can be achieved.

[0035] Generally, the terms "module," "unit," or "block" used in this specification refer to logic embodied in hardware or firmware, or a collection of software instructions. The modules, units, or blocks described in this specification may be implemented as software and / or hardware and may be stored in any type of non-transitory computer-readable medium or other storage device. In some embodiments, software modules / units / blocks may be compiled and linked into an executable program. It will be appreciated that software modules may be callable from other modules / units / blocks or themselves, and / or may be called in response to detected events or interrupts. Software modules / units / blocks configured to be executed on a computing device (e.g., processor 210 shown in FIG. 2 ) may be provided on a computer-readable medium. For example, a compact disc, digital video disc, flash drive, disk, 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 prior to execution). The software code herein may be stored in part or in full in a storage device of the computing device performing the operation and used in the operation of the computing device. The 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 computing device functions described in this specification may be implemented as software modules / units / blocks, but may be represented in hardware or firmware. Typically, the modules / units / blocks described herein refer to logical modules / units / blocks, which may be combined with other modules / units / blocks or divided into submodules / subunits / subblocks, although they are physical organizations or storage devices. This description may be applicable to a system, an engine, or a portion thereof.

[0036] 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 specification, the term "and / or" may include any one or more of the relevant listed items or any combination thereof.

[0037] These and other features and characteristics of the present specification, as well as the functions and methods of operation of the related structural elements, as well as the assembly of components and manufacturing economies, will become more apparent from the following description of the accompanying drawings, which form a part of this specification. However, it should be understood that the drawings are for illustration and description purposes only and are not intended to limit the scope of this specification. It should be understood that the drawings are not drawn to scale.

[0038] The flowcharts used in this specification illustrate operations performed by the systems described in some embodiments disclosed herein. It should be understood that the operations in the flowcharts may not be performed in order. Instead, the various steps may be performed in reverse order or simultaneously. Furthermore, one or more additional operations may be added to these flowcharts. One or more operations may also be deleted from the flowcharts.

[0039] Radiation therapy is the use of radiation to treat tumors. Due to the high energy of the radiation beam, while tumor cells are killed, normal cells are also affected. To minimize damage to normal tissue, a radiation therapy plan is necessary. Furthermore, to ensure accurate implementation of the radiation therapy plan, quality assurance testing of the radiation therapy plan is necessary. Quality assurance testing of radiation therapy plans can involve monitoring parameters involved in the patient's radiation therapy plan (e.g., dose distribution).

[0040] In current quality assurance testing of treatment plans, the measured dose distribution is measured using physical equipment (e.g., dose film, ionization chamber, electronic portal imaging device (EPID)). This is then compared with the calculated dose distribution determined by the treatment planning system (TPS). Gamma analysis is then used to determine the gamma pass rate, which is then used to determine whether the quality assurance test has passed. However, this method is often expensive and time-consuming, requires independent verification of each patient's radiotherapy plan, and it is difficult to determine the reason for a quality assurance test failure. Furthermore, in some online radiotherapy processes (e.g., online adaptive radiotherapy), actual measurement conditions for quality assurance testing are often unavailable.

[0041] This specification provides a method and system for quality assurance. Specifically, the system can obtain the status information of the medical device (for example, beam information corresponding to multiple dose rates, in-place accuracy information of at least one component of the medical device at multiple positions, operating error of at least one component of the medical device, and test results of the medical device executing a data set). The system can obtain the target plan of the target object (for example, features related to the complexity of the target plan, target flux map). The system can determine the prediction results (for example, predicted EPID image, gamma pass rate, dose distribution, reasons for failure of quality assurance test, target adjustment method) based on the status information and target plan of the medical device. For example, the parameter information of the state information of the medical device, the target plan of the target object and / or the dose model can be input into the quality assurance model, and the quality assurance model can output the prediction results. The system can determine whether the quality assurance test is passed based on the prediction results. In response to determining that the quality assurance test is passed, the system can control the medical device to perform the target operation (for example, radiotherapy operation) on the target object. Since the influence of multiple types of information such as medical equipment, target plan and dose model on the prediction results is taken into account at the same time, the rationality, accuracy and effectiveness of the quality assurance test can be improved. If the prediction results meet pre-set conditions, the quality assurance test is determined to have passed. Based on the target plan, the medical device can then be controlled to perform radiotherapy on the target patient, eliminating the need for the device to perform quality assurance testing on the target plan, thereby streamlining the radiotherapy process. Furthermore, based on the prediction results, the system can analyze the reasons for quality assurance test failures and provide recommended adjustments to improve the efficiency of the radiotherapy process.

[0042] FIG1 is a schematic diagram of an exemplary medical system according to some embodiments of the present specification. Medical system 100 may include a medical device 110, a network 120, one or more terminals 130, a processing device 140, and a storage device 150. The components in medical system 100 may be connected in various ways. By way of example only, medical device 110 may be connected to processing device 140 directly (as indicated by the dashed double-headed arrow connecting medical device 110 and processing device 140) or via network 120. As another example, storage device 150 may be connected to medical device 110 directly (as indicated by the dashed double-headed arrow connecting storage device 150 and medical device 110) or via network 120. As another example, terminal 130 may be connected to processing device 140 directly (as indicated by the dashed double-headed arrow connecting terminal 130 and processing device 140) or via network 120. As yet another example, terminal 130 may be connected to storage device 150 directly (as indicated by the dashed double-headed arrow connecting terminal 130 and storage device 150 ) or through network 120 .

[0043] Medical device 110 can image and / or treat an object (e.g., a target object). In some embodiments, the object can include a biological object and / or a non-biological object. For example, the object can include a specific part of the human body, such as the head, chest, abdomen, etc. For another example, the object can be a patient to be scanned by medical device 110.

[0044] In some embodiments, the medical device 110 may be a medical treatment device for disease treatment or research purposes. In some embodiments, the medical device 110 may include a single-modality device, such as an X-ray therapy device, a Co-60 teletherapy device, a medical electron accelerator, etc. In some embodiments, the medical device 110 may be a multimodal (e.g., dual-modality) device to obtain medical images related to an object and perform radiotherapy on the object. For example, the medical device 110 may include an image-guided radiation therapy (IGRT) device. For example, a CT-guided radiation therapy device, an MRI-guided radiation therapy device. The devices provided above are for illustrative purposes only and are not intended to limit the scope of the present application specification. As used in this specification, the term "imaging modality" or "modality" may refer to an imaging method or technology that collects, generates, processes and / or analyzes imaging information of an object.

[0045] In some embodiments, the medical device 110 may include a radiotherapy component, such as a treatment head. The treatment head is connected to a gantry. In some embodiments, the treatment head can move with the movement (e.g., rotation) of the gantry. The treatment head may include a target, a therapeutic radiation source, and a beam limiter. The therapeutic radiation source can emit a radiation beam toward the subject. In some embodiments, the therapeutic radiation source can be a radiation source for emitting beams in the megavolt energy range (i.e., the energy of the beam is greater than 1 megavolt), or a radiation source for emitting beams in the kilovolt energy range (i.e., the energy of the beam is greater than 1 kilovolt). The beam limiter may include a beam limiting aperture, a primary collimator, and a secondary collimator. The secondary collimator may include a multi-leaf collimator and a tungsten gate, etc. The multi-leaf collimator may include multiple leaves. In some embodiments, the positions of the multiple leaves of the multi-leaf collimator and / or the tungsten gate are used to form a radiation area, i.e., a radiation field. In some embodiments, the multiple leaves can be driven by one or more drive components (e.g., a motor) to move to specific positions to change the shape of the radiation field.

[0046] In some embodiments, the medical device 110 may include a radiotherapy auxiliary device, such as an electronic portal imaging device (EPID). The electronic portal imaging device can generate images of the object before, during and / or after treatment. The electronic portal imaging device may include a detector for detecting radiation rays (e.g., X-rays, gamma rays) emitted from a therapeutic radiation source. In some embodiments, the detector may include one or more detection units. The detection unit may include a scintillation detector (e.g., a cesium iodide detector, a gadolinium oxysulfide detector), a gas detector, etc. The detection unit may include a single-row detector or a multi-row detector.

[0047] In some embodiments, the medical device 110 may further include an imaging device. The imaging device may include one or a combination of a computed tomography (CT) device, an ultrasound imaging component, a fluoroscopic imaging component, a magnetic resonance imaging (MRI) device, a single photon emission computed tomography (SPECT) device, a positron emission tomography (PET) device, and the like. In some embodiments, the imaging device may include an imaging component, such as an imaging radiation source and a detector. In some embodiments, the imaging radiation source and the therapeutic radiation source may be integrated into a single radiation source for imaging and / or treating a subject.

[0048] The network 120 may comprise any suitable network that can facilitate the exchange of information and / or data for the medical system 100. In some embodiments, one or more components of the medical system 100 (e.g., the medical device 110, the terminal 130, the processing device 140, the storage device 150) can communicate information and / or data with one or more other components of the medical system 100 via the network 120. For example, the processing device 140 can obtain image data from the medical device 110 via the network 120. As another example, the processing device 140 can obtain user instructions from the terminal 130 via the network 120. The network 120 may be and / or include a public network (e.g., the Internet), a private network (e.g., a local area network (LAN), a wide area network (WAN)), a wired network (e.g., a wireless LAN), Ethernet, a wireless network (e.g., an 802.11 network, a Wi-Fi network), a cellular network (e.g., a Long Term Evolution (LTE) 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. By way of example only, the network 120 may include a cable network, a wired network, a fiber optic network, a telecommunications network, an intranet, a wireless local area network (WLAN), a metropolitan area network (MAN), a public switched telephone network (PSTN), a Bluetooth network, a ZigBee network, a near field communication (NFC) network, or the like, 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 wired and / or wireless network access points such as base stations and / or Internet exchange points, and one or more components of the medical system 100 may connect to the network 120 via the wired and / or wireless access points to exchange data and / or information.

[0049] Terminal 130 may include a mobile device 130-1, a tablet computer 130-2, a laptop computer 130-3, or the like, or any combination thereof. In some embodiments, mobile device 130-1 may include a smart home device, a wearable device, a smart mobile device, a virtual reality device, an augmented reality device, or the like, or any combination thereof. By way of example only, terminal 130 may include the mobile device shown in FIG3 . In some embodiments, smart home devices may include smart lighting devices, smart appliance control devices, smart monitoring devices, smart televisions, smart cameras, intercoms, or the like, or any combination thereof. In some embodiments, wearable devices may include bracelets, footwear, glasses, helmets, watches, clothing, backpacks, smart accessories, or the like, 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, or the like, 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, or the like, or any combination thereof. In some embodiments, one or more terminals 130 may be part of processing device 140. In some embodiments, one or more terminals 130 may be part of the medical device 110 .

[0050] The processing device 140 can process data and / or information obtained from the medical device 110, the terminal 130, and / or the storage device 150. For example, the processing device 140 can obtain status information of the medical device 110 and a target plan for a target object. As another example, the processing device 140 can determine a prediction result based on the status information and the target plan of the medical device. As another example, the processing device 140 can determine whether a quality assurance test has passed based on the prediction result. As another example, in response to determining that the quality assurance test has passed, the processing device 140 can control the medical device to perform a target operation on the target object. In some embodiments, the processing device 140 can be a single server or a server group. The server group can be centralized or distributed. In some embodiments, the processing device 140 can be a local component or a remote component relative to one or more other components of the medical system 100. For example, the processing device 140 can access information and / or data stored in the medical device 110, the terminal 130, and / or the storage device 150 via the network 120. As another example, the processing device 140 can be directly connected to the medical device 110, the terminal 130, and / or the storage device 150 to access stored information and / or data. In some embodiments, the processing device 140 can be implemented on a cloud platform. By way of example only, the cloud platform can include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, or the like, or any combination thereof. In some embodiments, the processing device 140 can be implemented by a computing device 200 having one or more components as shown in FIG. 2 .

[0051] The storage device 150 can store data, instructions, and / or any other information. In some embodiments, the storage device 150 can store data obtained from the terminal 130 and / or the processing device 140. In some embodiments, the storage device 150 can store data and / or instructions that the processing device 140 can execute or use to execute the exemplary methods described in this application specification. In some embodiments, the storage device 150 can include a mass storage device, a removable storage device, a volatile read-write memory, a read-only memory (ROM), or the like, or any combination thereof. Exemplary mass storage devices can include magnetic disks, optical disks, solid-state drives, etc. Exemplary removable storage devices can include flash drives, floppy disks, optical disks, memory cards, compact disks, magnetic tapes, etc. Exemplary volatile read-write memory devices can include random access memory (RAM). Exemplary RAM devices can include dynamic random access memory (DRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), static random access memory (SRAM), thyristor random access memory (T-RAM), and zero-capacitance random access memory (Z-RAM). Exemplary ROMs may include mask ROM (MROM), programmable ROM (PROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), compact disc ROM (CD-ROM), and digital versatile disk ROM. 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, or any combination thereof.

[0052] In some embodiments, the storage device 150 can be connected to the network 120 to communicate with one or more other components of the medical system 100 (e.g., the processing device 140, the terminal 130). One or more components of the medical system 100 can access data or instructions stored in the storage device 150 via the network 120. In some embodiments, the storage device 150 can be directly connected to or communicate with one or more other components of the medical system 100 (e.g., the processing device 140, the terminal 130). In some embodiments, the storage device 150 can be part of the processing device 140.

[0053] It should be noted that the above description is provided for illustrative purposes only and is not intended to limit the scope of this application specification. For those of ordinary skill in the art, under the guidance of this application specification, various changes and modifications can be made. The features, structures, methods and other features of the exemplary embodiments described in this application specification can be combined in various ways to obtain additional and / or alternative exemplary embodiments. However, these changes and modifications will not deviate from the scope of this application specification.

[0054] FIG2 is a diagram of an exemplary computing device on which a processing device may be implemented according to some embodiments of the present specification. As shown in FIG2 , computing device 200 may include a processor 210 , a memory 220 , an input / output 230 , and a communication port 240 .

[0055] The processor 210 can execute computer instructions (e.g., program code) and perform the functions of the processing device 140 according to the techniques described in this specification. Computer instructions may include, for example, routines, programs, objects, components, data structures, processes, modules, and functions that perform the specific functions described in this specification. For example, the processor 210 can process data or information obtained from the medical device 110, the storage device 150, the terminal 130, and / or any other component of the medical 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, or a combination thereof.

[0056] For illustration purposes only, only one processor is described in the computing device 200. However, it should be noted that the computing device 200 disclosed in this specification may also include multiple processors. Therefore, the operations and / or method steps disclosed in this specification that are performed by one processor may also be performed jointly or separately by multiple processors. For example, if in this specification, the processor of the computing device 200 performs operations A and B, it should be understood that operations A and B may also be performed jointly or separately by two or more different processors in the computing device 200 (for example, the first processor performs operation A, the second processor performs operation B, or the first processor and the second processor jointly perform operations A and B).

[0057] The memory 220 can store data / information obtained from the medical device 110, the storage device 150, the terminal 130, and / or any other component of the medical system 100. In some embodiments, the memory 220 can include mass storage, removable storage, volatile read-write memory, read-only memory, or the like, or any combination thereof. For example, the mass storage can include a magnetic disk, an optical disk, a solid-state drive, or the like. The removable storage can include a flash drive, a floppy disk, an optical disk, a memory card, a compact disk, a magnetic tape, or the like. The volatile read-write memory can include random access memory (RAM). RAM can include dynamic RAM (DRAM), double data rate synchronous dynamic RAM (DDR SDRAM), static RAM (SRAM), thyristor RAM (T-RAM), and zero-capacitor RAM (Z-RAM). ROM can 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, or the like. In some embodiments, the memory 220 can store one or more programs and / or instructions for executing the exemplary methods described herein.

[0058] The input / output 230 can input and / or output signals, data, information, and the like. In some embodiments, the input / output 230 can enable a user to interact with the processing device 140. In some embodiments, the input / output 230 can include input devices and output devices. Exemplary input devices can include a keyboard, a mouse, a touch screen, a microphone, or the like, or a combination thereof. Exemplary output devices can include a display device, a speaker, a printer, a projector, or the like, or a combination thereof. Exemplary display devices can include a liquid crystal display (LCD), a light emitting diode (LED)-based display, a flat panel display, a curved screen, a television device, a cathode ray tube (CRT), a touch screen screen, or the like, or a combination thereof.

[0059] 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 medical device 110, the storage device 150 and / or the terminal 130. The connection can be a wired connection, a wireless connection, any other communication connection that can realize data transmission and / or reception, and / or a combination of these connections. The wired connection may include, for example, an electrical cable, an optical cable, a telephone line, etc., or any combination thereof. The wireless connection may include, for example, Bluetooth, Wi-Fi, WiMax, wireless LAN Zifeng, a mobile network (e.g., 3G, 4G, 5G), etc., or a combination thereof. In some embodiments, the communication port 240 can be and / or include a standardized communication port, for example, RS232, RS485, etc. In some embodiments, the communication port 240 can be a specially designed communication port. For example, the communication port 240 can be designed according to the Digital Imaging and Communications in Medicine (DICOM) protocol.

[0060] FIG3 is a diagram of exemplary hardware and / or software components of an exemplary mobile device on which a terminal may be implemented according to some embodiments of the present specification. As shown in FIG3 , mobile device 300 may include a communication platform 310, a display 320, a graphics processing unit 330, a central processing unit 340, an input / output 350, a memory 360, and a 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, a mobile operating system 370 (e.g., iOS 11) may be included. TM 、Android TM 、Windows Phone TM ) and one or more applications 380 are loaded from storage 390 into memory 360 for execution by central processing unit 340. Applications 380 may include a browser or any other suitable mobile application for receiving and rendering information related to medical system 100 or other information from processing device 140. User interaction with the information may be enabled via input / output 350 and provided to processing device 140 and / or other components of medical system 100 via network 120.

[0061] To implement the various modules, units, and their functions described in this specification, a computer hardware platform can be used as the hardware platform for one or more components described in this specification. A computer with user interface elements can be used as a personal computer (PC) or any other type of workstation or terminal device. If the computer is appropriately programmed, it can also be used as a server.

[0062] 4 is a block diagram of an exemplary processing device according to some embodiments of the present disclosure, wherein the processing device 140 may include a first acquisition module 410 , a second acquisition module 420 , a determination module 430 , and a control model 440 .

[0063] The first acquisition module 410 can acquire status information of a medical device (e.g., medical device 110). The status information of the medical device can include at least one of beam information corresponding to multiple dose rates, positioning accuracy information of at least one component of the medical device at multiple positions, or operational error of at least one component of the medical device. Further description of acquiring status information of a medical device can be found elsewhere in this specification (e.g., step 510 in FIG. 5 ).

[0064] The second acquisition module 420 can acquire a target plan for the target object. The target plan may include radiotherapy parameters involved in the radiotherapy process. In some embodiments, the second acquisition module 420 can determine features and / or a target flux map related to the complexity of the target plan based on the target plan. More description of acquiring status information of the medical device can be found elsewhere in this specification (e.g., step 520 in Figure 5).

[0065] The determination module 430 may determine a prediction result based on the medical device's status information and the target plan. In some embodiments, the determination module 430 may determine whether the medical device's status information satisfies a preset condition. Further description regarding determining whether the medical device's status information satisfies the preset condition can be found elsewhere in this specification (e.g., step 530 in FIG. 5 ). In response to determining that the medical device's status information satisfies the preset condition, the determination module 430 may determine a prediction result based on the medical device's status information and the target plan. Further description regarding determining a prediction result can be found elsewhere in this specification (e.g., step 540 in FIG. 5 ). In some embodiments, the determination module 430 may determine whether a quality assurance test passes based on the prediction result. Further description regarding determining whether a quality assurance test passes can be found elsewhere in this specification (e.g., step 550 in FIG. 5 ). In response to determining that a quality assurance test fails, the determination module 430 may adjust the target plan or perform a quality assurance test on the target subject. Further description regarding controlling the medical device to execute the target plan on the target subject can be found elsewhere in this specification (e.g., step 570 in FIG. 5 ).

[0066] The control module 440 can control one or more components of the medical device. In some embodiments, in response to determining that the quality assurance test passes, the control module 440 can control the medical device to execute the target plan on the target subject. Further description of controlling the medical device to execute the target plan on the target subject can be found elsewhere in this specification (e.g., step 560 in FIG. 5 ).

[0067] It should be noted that the above description of the processing device 140 in this specification is provided for illustrative purposes only and is not intended to limit the scope of this specification. For those skilled in the art, various changes and modifications can be made based on the description of this specification. However, these changes and modifications do not depart from the scope of this specification. For example, the processing device 140 may further include a storage module (not shown in the figure) for data storage. For another example, the first acquisition module 410 and the second acquisition module 420 may be integrated into one module. For another example, the processing device 140 may further include a training model (not shown in the figure) for training a quality assurance model.

[0068] FIG5 is a flowchart of an exemplary process of a quality assurance method according to some embodiments of the present specification. In some embodiments, at least a portion of process 500 can be performed by processing device 140 (e.g., implemented in computing device 200 shown in FIG2 ). For example, process 500 can be stored in a storage device (e.g., storage device 150, memory 220, memory 390) in the form of instructions (e.g., application programs) and called and / or executed by processing device 140 (e.g., processor 210 shown in FIG2 , CPU 340 shown in FIG3 , or one or more modules in processing device 140 shown in FIG4 ). The operations of the process shown below are for illustrative purposes only. In some embodiments, process 500 can be completed using one or more additional operations not described and / or without one or more operations discussed. In addition, the order of the operations of process 500 shown in FIG5 and described below is not intended to be limiting.

[0069] In 510 , the processing device 140 (eg, the first acquisition module 410 ) may acquire status information of a medical device (eg, the medical device 110 ).

[0070] In some embodiments, the status information of a medical device can reflect the status of the medical device. For example, the status information of the medical device can be expressed through machine operating parameters; the status of the medical device can also be predicted through the results of the medical device implementing a radiotherapy plan. In some embodiments, the status information of the medical device can include beam and dose information, positioning accuracy information of at least one component of the medical device (e.g., gantry, table, radiotherapy radiation source, beam limiter), operating error of at least one component of the medical device, results of executing a dataset using the medical device, or any combination thereof. In some embodiments, at least one of the following can be obtained by analyzing the results of executing a dataset using the medical device. The beam and dose information can include beam flatness, beam symmetry, beam output correction, beam linearity, the relationship between beam output consistency and dose rate, or any combination thereof. Beam output correction can refer to machine MU correction. For example, after beam output correction, the center dose at the maximum dose depth in a 10 cm × 10 cm field can be calibrated to 1 MU = 1 cGy. In some embodiments, the beam and dose information may include beam information at multiple dose rates. For example, the beam and dose information may include beam flatness, beam symmetry, beam linearity, beam output consistency, etc. at multiple dose rates. The multiple dose rates may be set by a user of the medical system 100 (e.g., a physician, a technician) or by one or more components of the medical system 100 (e.g., the processing device 140) according to different circumstances.

[0071] The positioning accuracy information of at least one component of the medical device may include the accuracy of the gantry rotation isocenter, the accuracy of the beam limiter rotation isocenter, the positioning accuracy of the scanning bed, the positioning accuracy information of the beam limiter, or the like, or any combination thereof. The positioning accuracy information of the beam limiter may include the positioning accuracy of the multi-leaf collimator blades, the accuracy of the blade motion speed, the repeatability of the blade positioning, or the like, or any combination thereof. In some embodiments, the positioning accuracy information of at least one component of the medical device may include the positioning accuracy information of the at least one component of the medical device at multiple positions. The multiple positions may include multiple gantry angles, multiple beam limiter positions (e.g., primary collimator positions, secondary collimator positions), multiple beam limiter angles (e.g., primary collimator angles, secondary collimator angles), or the like, or any combination thereof. For example, the positioning accuracy information of the beam limiter may include the positioning accuracy of the beam limiter at multiple gantry angles. The multiple positions may be set by a user of the medical system 100 (e.g., a doctor, a technician) or by one or more components of the medical system 100 (e.g., the processing device 140) according to different circumstances. For example, multiple rack angles can be set to 30°, 60°, 120°, 180°, 240°, 300°, etc.

[0072] Since the status information of medical equipment under different test conditions is usually different, for example, the beam information under different dose rates and the beam limiter positioning accuracy information under different gantry angles are usually different, therefore, fully considering the status information of medical equipment under different test conditions during quality assurance testing can make the results of quality assurance testing more accurate and effective.

[0073] The operational errors of the components of the medical device may include systematic errors and / or random errors. The systematic error of a component may refer to the deviation of the component as a whole. For example, the systematic error of a multi-leaf collimator may refer to the overall offset of all the leaves in the multi-leaf collimator (for example, all the leaves in the multi-leaf collimator are offset a certain distance in a certain direction as a whole). The random error of a component may refer to the deviation of a certain sub-component in the component. For example, the offset of each leaf in the multi-leaf collimator may be different (for example, the offset of the leaves of the multi-leaf collimator is generally Gaussian distributed), and the random error of the multi-leaf collimator may refer to the offset of a certain leaf in the multi-leaf collimator (for example, the leaf is offset a certain distance in a certain direction).

[0074] In some embodiments, the medical device status information may include operational errors of one or more components of the medical device at one or more positions (e.g., multiple gantry angles, multiple beam limiter positions, multiple beam limiter angles). For example, the medical device status information may include systematic errors and random errors of a multi-leaf collimator at multiple gantry angles, systematic errors and random errors of a multi-leaf collimator at multiple beam limiter positions and / or angles, and the like, or any combination thereof.

[0075] In some embodiments, the processing device 140 can obtain operating information of the medical device. For example, the processing device 140 can obtain historical quality assurance test data of the medical device. Historical quality assurance tests may include quality assurance tests such as daily inspections, weekly inspections, monthly inspections, and annual inspections of the medical device. Historical quality assurance test data may include historical log files and historical EPID images. In some embodiments, the operating information of the medical device may be stored in a storage device (e.g., storage device 150) or an external storage device of the medical system 100, and the processing device 140 may obtain the operating information of the medical device from the storage device (e.g., storage device 150) or the external storage device of the medical system 100. The processing device 140 can determine the operating error of at least one component of the medical device (e.g., a gantry, a radiotherapy radiation source, a beam limiter) based on the operating information of the medical device.

[0076] In some embodiments, the processing device 140 can obtain actual operating information and expected operating information of at least one component of the medical device based on the operating information of the medical device. The expected operating information of the component can be the operating information (e.g., expected position, expected movement speed, expected movement acceleration, expected movement distance, expected movement direction) of the component set by the user of the medical system 100 (e.g., doctor, technician) through the control device of the medical device (e.g., terminal 130). The processing device 140 can determine the operating error of at least one component of the medical device based on the actual operating information and the expected operating information. For example, the processing device 140 can read the expected operating parameters of the component and the actual operating parameters of the component input by the user from the historical log file of the medical device. For another example, a test device (e.g., position encoder, velocity sensor, acceleration sensor) can be installed on a component of the medical device (e.g., scanning bed, gantry) to detect the actual operating information of the component (e.g., actual position, actual movement speed, actual movement acceleration, actual movement distance, actual movement direction). For example only, a user may determine that a desired gantry rotation angle is 10° based on a radiotherapy plan, and input the desired gantry rotation angle of 10° into the medical device to control the rotation of the gantry. However, due to an operational error of the gantry, the actual gantry rotation angle is 12°. The processing device 140 may determine that the operational error of the gantry is 2° (12°-10°=2°).

[0077] In some embodiments, the processing device 140 may determine the operational error of at least one component of the medical device based on the operational information of the medical device and according to a first error model. The first error model is used to determine the operational error of at least one component of the medical device based on the operational information of the medical device. For example, the operational information of the medical device may be input into the first error model, and the first error model may output the operational error of at least one component of the medical device. In some embodiments, the processing device 140 may determine the operational error of at least one component of the medical device based on a second error model based on simulated images and / or test images, and a radiotherapy plan for a subject (e.g., a patient). The simulated image may be an EPID image determined based on the subject's radiotherapy plan and according to a data simulation algorithm. The test image may be an EPID image acquired by an imaging device (e.g., an electronic portal imaging device) based on the subject's radiotherapy plan. For example, the simulated image and / or test image, as well as the subject's radiotherapy plan, may be input into the second error model, and the second error model may output the operational error of at least one component of the medical device. The first error model and / or the second error model may be machine learning models. For example, the first error model and / or the second error model can be a convolutional neural network model (CNN), a recurrent neural network model (RNN), a deep neural network model (DNN), a fully convolutional neural network (FCN) model, a generative adversarial network (GAN) model, etc.

[0078] As medical equipment is used for a longer time and is affected by the usage environment, its operating status usually changes. Different operating statuses of medical equipment have a significant impact on quality assurance test results. For example, for the same radiotherapy plan, when at least one of the medical equipment's beam and dose information, component placement accuracy, or operating error changes, the predicted results may be significantly different. Therefore, fully considering the current operating status of the medical equipment during quality assurance testing can make the quality assurance test results more accurate and effective.

[0079] In some embodiments, the processing device 140 can acquire at least one data set. A data set can refer to a collection of data related to radiotherapy parameters. In some embodiments, the at least one data set can be determined based on a candidate plan, the ultimate performance of at least one component of the medical device, the calculated limit of a dose model, or the like, or any combination thereof. In some embodiments, the at least one data set can include a first data set, a second data set, a third data set, or the like, or any combination thereof. In some embodiments, the data set can be composed of multiple radiotherapy plans, which can include typical clinical plans. In some embodiments, the data set can also include radiotherapy plans used to examine the ultimate performance of components and / or the calculated limit of a dose model.

[0080] In some embodiments, the processing device 140 may determine a first data set based on at least one first candidate plan. The value of the first candidate parameter in the at least one first candidate plan is within a first preset range. The first candidate plan may be a historical radiotherapy plan for a candidate subject. The candidate subject may include a patient who has undergone historical radiotherapy. The candidate parameters (e.g., first candidate parameters, second candidate parameters) may include radiotherapy parameters involved in the radiotherapy process (e.g., dose rate of the radiation source, radiation duration, gantry angle, gantry motion speed, gantry motion acceleration, beam limiter angle, beam limiter rotation speed, parameters of the leaves of the multi-leaf collimator, position and / or angle of the scanning table, number of fields (or sub-fields), field (or sub-field) shape, field (or sub-field) area, field (or sub-field) perimeter, field angle, etc., as described in step 520). In some embodiments, the first preset range may be set by a user (e.g., a doctor, a technician) based on experience. In some embodiments, the first preset range may be determined based on the values ​​of the radiotherapy parameters in multiple historical radiotherapy plans. For example, taking the number of radiation fields as an example of a radiotherapy parameter, if the number of radiation fields in historical radiotherapy plans that is greater than a preset threshold (e.g., 90%, 95%, 99%) is between 1 and 5, then the first preset range can be set to 1 to 5. In other words, the first data set can be considered as a collection of data determined based on radiotherapy plans that are more commonly used in clinical practice, that is, the range of the values ​​of the radiotherapy parameters in the first data set is within the range of the values ​​of the corresponding radiotherapy parameters in most historical radiotherapy plans.

[0081] In some embodiments, the processing device 140 may directly use the values ​​of the radiotherapy parameters in the first candidate plan as the values ​​of the radiotherapy parameters in the first data set. In some embodiments, the processing device 140 may adjust the values ​​of the radiotherapy parameters in the first candidate plan and use the adjusted values ​​of the radiotherapy parameters as the values ​​of the radiotherapy parameters in the first data set. In some embodiments, the processing device 140 may input multiple first candidate plans into a parameter extraction model, the parameter extraction model may extract the values ​​of the radiotherapy parameters in the multiple first candidate plans, and the processing device 140 may use the extracted values ​​of the radiotherapy parameters as the values ​​of the radiotherapy parameters in the first data set. In some embodiments, the processing device 140 may input multiple first candidate plans into a parameter extraction model, and the parameter extraction model may output a first data set. The parameter extraction model may refer to a machine learning model or algorithm used to extract the values ​​of the radiotherapy parameters in the candidate plans (e.g., the first candidate plan, the second candidate plan) or determine the data sets (e.g., the first data set, the second data set) based on the candidate plans.

[0082] In some embodiments, the processing device 140 may determine a second dataset based on at least one second candidate plan. The value of a second candidate parameter in the second candidate plan is outside a second preset range. The second candidate plan may be a historical radiotherapy plan for the candidate subject. The first preset range and the second preset range may be the same or different. In some embodiments, the second preset range may be set by a user (e.g., a physician or technician) based on experience. In some embodiments, the second preset range may also be determined based on the values ​​of the radiotherapy parameter in multiple historical radiotherapy plans. For example, using the number of fields as an example, if the number of fields in historical radiotherapy plans that exceeds a preset threshold (e.g., 90%, 95%, 99%) is between 1 and 5, then the second preset range may be set to be greater than 5. For another example, the second preset range may be determined based on the values ​​of the radiotherapy parameter in certain special types of radiotherapy plans (e.g., large-field radiotherapy plans). For another example, if the value of a certain radiotherapy parameter in a historical radiotherapy plan based on a certain radiotherapy objective needs to be set higher or lower (e.g., a smaller sub-field area, a sub-field with a greater degree of deviation from the beam center), the value of the radiotherapy parameter in the historical radiotherapy plan may be used as the value of the radiotherapy parameter in the second dataset. That is, the second data set can be considered as a set of data determined based on radiotherapy plans that are not common in clinical practice, that is, the range of values ​​of the radiotherapy parameters in the second data set is outside the range of values ​​of the corresponding radiotherapy parameters in most historical radiotherapy plans. The second data set is determined in a similar manner to the first data set and will not be described in detail here. For example, the processing device 140 can directly use the values ​​of the radiotherapy parameters in the second candidate plan as the values ​​of the radiotherapy parameters in the second data set. For another example, the processing device 140 can determine the values ​​of the radiotherapy parameters in the second data set based on a parameter extraction model based on multiple second candidate plans.

[0083] In some embodiments, the processing device 140 may determine a third data set based on the extreme performance of at least one component of the medical device and / or parameter information of the dose model (e.g., the calculation limit of the dose model). For example, the third data set may include the extreme performance parameters of at least one component of the medical device and / or parameters related to the calculation limit of the dose model. The extreme performance of the medical device may include the maximum movement speed, maximum movement acceleration, extreme movement position, etc. of the components of the medical device (e.g., radiation source, gantry, scanning table, leaves of a multi-leaf collimator), or any combination thereof. Since the performance of each component of the medical device cannot exceed its extreme performance, the values ​​of the parameters related to the radiotherapy process in the third data set need to be reasonably set according to the extreme performance of the components of the medical device.

[0084] A dose model includes an algorithm or model that can simulate and calculate dose-related results (e.g., EPID images, dose distribution) based on radiotherapy parameter values. In some embodiments, the parameter information of the dose model can include the type of dose model (e.g., the type of dose algorithm), parameters of the dose model (e.g., parameters in the dose algorithm), etc., or any combination thereof. The dose model (or dose algorithm) can include a Monte Carlo dose model, a pencil beam algorithm, a convolution algorithm, a machine learning algorithm, etc., or any combination thereof.

[0085] In some embodiments, the parameter information of the dose model may include parameters of a model of one or more components of the medical device. For example, the parameter information of the dose model may include parameters of an accelerator model (e.g., a beam model). The accelerator model can be used to determine the energy (absorbed dose) deposited by the beam in the target area and organs at risk. In some embodiments, the parameters of the accelerator model may include relevant parameters (e.g., position parameters) of components such as a primary collimator, a uniformizer, and a secondary collimator. For example, the parameter information of the dose model may include multi-leaf collimator deviation, multi-leaf collimator transmittance, the size of the leaf-tip structure of the multi-leaf collimator blades, the width and transmittance of the tongue-and-groove structure of the multi-leaf collimator blades, etc., or any combination thereof.

[0086] The use of dose models can simplify the dose calculation process and improve the efficiency of dose calculation. However, the use of different types of dose models or different dose model parameters will have a greater impact on the quality assurance test results. For example, for the same radiotherapy plan, when different dose model parameters are used, the predicted results may be significantly different. Therefore, fully considering the parameter information of the dose model during quality assurance testing can make the quality assurance test results more accurate and effective.

[0087] In some embodiments, the calculation limit of the dose model can reflect the calculation capability of the dose model. If the calculation capability of the dose model is poor (for example, the calculation accuracy is low), the pass rate of the dose model in the test of the calculation limit will be low. Therefore, the test of the calculation limit can reflect the calculation capability of the dose model. For example, the calculation limit of the dose model can include the calculation accuracy of the dose model for parameter values ​​that are prone to calculation deviations. For example, since the rays within the field are usually unevenly distributed, the dose rate of the rays near the edge of the field will be low, and the non-uniform distribution of the rays can easily lead to errors in the dose calculation (especially the dose calculation at the field boundary). Therefore, the accuracy of the dose model for the dose calculation at the field boundary can reflect the calculation capability of the dose model.

[0088] In some embodiments, at least one data set may also include other types of data sets. For example, a user of the medical system 100 (e.g., a doctor, a technician) may set the values ​​of the radiotherapy parameters in the data set based on experience. For another example, a user of the medical system 100 (e.g., a doctor, a technician) may set a range of values ​​for each radiotherapy parameter in the data set, and the processing device 140 may randomly set the values ​​of the radiotherapy parameters based on the range of values ​​of the radiotherapy parameters to generate a data set. For another example, the processing device 140 may randomly select a historical radiotherapy plan from a database and use the values ​​of the radiotherapy parameters in the selected historical radiotherapy plan as the values ​​of the radiotherapy parameters in the data set.

[0089] In some embodiments, the processing device 140 can determine actual test results based on at least one data set. For example, the values ​​of the radiotherapy parameters in the first data set can be executed when the gantry angle is between 0° and 30°, the values ​​of the radiotherapy parameters in the second data set can be executed when the gantry angle is between 30° and 60°, and the values ​​of the radiotherapy parameters in the third data set can be executed when the gantry angle is between 60° and 120°. In some embodiments, the actual test results can be determined based on at least one data set by using medical equipment and / or testing devices. The actual test results may include images, dose distributions, etc., or any combination thereof. The dose distribution may refer to the spatial distribution of energy deposited in an object by radioactive particles such as photons or charged particles. In some embodiments, the dose distribution may include the dose delivered to one or more parts of the object and / or the absorbed dose absorbed by one or more parts of the object. In some embodiments, the testing device may include an imaging device (e.g., an electronic portal imaging device). The actual test results may include test images acquired by the imaging device (e.g., an electronic portal imaging device) based on at least one data set. For example, a radiotherapy process can be performed according to the parameters of the radiotherapy plan in at least one data set, and the electronic portal imaging device receives the irradiation of the portal and obtains a portal image. At this time, the test image can be a portal image. In some embodiments, the processing device 140 can reconstruct the acquired portal image into a dose distribution map. In some embodiments, the test device can include a third-party device (for example, a phantom, a dose measurement device). The actual test results may include a measured dose distribution detected by the dose measurement device based on the data set. For example, the phantom can be placed on a scanning table, and a radiotherapy process can be performed on the phantom according to the radiotherapy parameter values ​​in the data set, and the measured dose distribution in the phantom can be detected by the dose measurement device. For another example, a radiotherapy process can be performed on the phantom according to the parameters of the radiotherapy plan in at least one data set, and a test image of the phantom (for example, the acquired portal image or the converted dose distribution image) can be collected by the electronic portal imaging device.

[0090] In some embodiments, the processing device 140 can determine the predicted test results based on at least one data set. The predicted test results may include images, dose distributions, etc. or any combination thereof. In some embodiments, the processing device 140 can obtain simulated scan data based on at least one data set according to a dose model or other data simulation algorithm. The processing device 140 can generate simulated images based on the simulated scan data. In some embodiments, the processing device 140 can obtain simulated dose distributions based on at least one data set according to a dose model. In some embodiments, the predicted test results may also include a planned dose distribution. The planned dose distribution may refer to the expected dose distribution. The planned dose distribution may be set by a user of the medical system 100 (e.g., a technician, a doctor) or by a component of the medical device 110 (e.g., the processing device 140) according to actual conditions.

[0091] Furthermore, the processing device 140 can determine a dataset execution result for the medical device based on the actual test results and the predicted test results. The dataset execution result can include the difference between the actual test results and the predicted test results. For example, the processing device 140 can determine the difference between the measured dose distribution and the simulated dose distribution (or the planned dose distribution). In some embodiments, by executing at least one dataset using the medical device, the difference between the actual test results and the predicted test results obtained can reflect the status of the medical device. For example, the smaller the difference between the actual test results and the predicted test results obtained when the medical device executes at least one dataset, the better the status of the medical device and the higher the accuracy of the radiotherapy plan executed using the medical device. Furthermore, by executing different types of datasets using the medical device, the difference between the actual test results and the predicted test results obtained can reflect different types of medical device status information. For example, executing the first dataset can reflect the accuracy of the medical device in executing a more common radiotherapy plan in clinical practice, while executing the second and / or third datasets can reflect the accuracy of the medical device in executing less common radiotherapy plans in clinical practice. In some embodiments, the same data set (e.g., a second data set, a third data set) can be executed multiple times using a medical device to obtain multiple execution results (e.g., multiple differences between actual test results and predicted test results). The differences between the multiple execution results can reflect changes in the status information of the medical device. For example, during factory inspection of a medical device, a first execution result can be determined by executing a third data set using the medical device. After the medical device has been used for a period of time, the third data set can be executed again using the medical device to determine a second execution result. The difference between the first execution result and the second execution result can reflect the degree of change between the current state of the medical device and the state at the time of factory inspection.

[0092] In some embodiments, the first data set and / or the second data set may include multiple data subsets. The multiple data subsets are configured to correspond to different types of radiotherapy plans. The different types of radiotherapy plans are classified based on the site receiving radiotherapy, the type of target volume (e.g., tumor), and / or the type of radiotherapy technique (e.g., two-dimensional conformal radiotherapy, three-dimensional conformal radiotherapy, intensity-modulated radiation therapy, image-guided radiotherapy, biologically guided radiotherapy, dose-guided radiotherapy, online adaptive radiotherapy, etc.).

[0093] In 520 , the processing device 140 (eg, the second acquisition module 420 ) may acquire a target plan for the target object.

[0094] The target object may refer to a patient to be treated with radiotherapy. The target plan (e.g., a target radiotherapy plan) may indicate how to execute the radiotherapy plan on the target object. For example, the target plan may include radiotherapy parameters involved in the radiotherapy process. For example, the radiotherapy parameters may include the dose rate of the radiation source (i.e., MUs / min), the duration of radiation, the gantry angle, the gantry movement speed, the gantry movement acceleration, the beam limiter angle, the beam limiter rotation speed, the parameters of the leaves of the multi-leaf collimator (e.g., position, movement speed, movement acceleration, movement direction), the position and / or angle of the scanning table, the number of fields (or sub-fields), the shape of the fields (or sub-fields), the area of ​​the fields (or sub-fields), the perimeter of the fields (or sub-fields), the angle of the fields, the angle of the fields, etc., or any combination thereof. In some embodiments, the target plan may include an intensity modulated radiotherapy (IMRT) plan, a dynamic rotation intensity modulated radiotherapy (uARC) plan, an image-guided radiotherapy plan, a biologically guided radiotherapy plan, a dose-guided radiotherapy plan, an online adaptive radiotherapy plan, etc., or any combination thereof.

[0095] In some embodiments, a sub-field is a small field with a certain shape that can pass through rays, and the superposition of the shapes of multiple sub-fields (which can be called a radiation field) can match the shape of the target area. The number of sub-fields can be determined according to the type of radiotherapy equipment, the position, area, shape of the target area, etc. The shape of the sub-field can be achieved by the position of the leaves and / or tungsten gate in the multi-leaf collimator. During radiotherapy, sub-fields of different shapes and areas can be formed by changes in the leaf spacing, movement direction, movement speed, etc. in the multi-leaf collimator. The dose rate of the radiation source can refer to the intensity of the particle beam or beam generated by the radiation source passing through the sub-field to irradiate the target area per unit time. The radiation duration can refer to the duration of the particle beam or beam generated by the radiotherapy equipment (for example, a therapeutic radiation source) passing through the sub-field to irradiate the target area. In some embodiments, the dose distribution can be determined based on the dose rate and radiation duration at a specific angle. The radiation field angle can refer to the incident angle of the particle beam or beam generated by the radiotherapy equipment (for example, a therapeutic radiation source) to irradiate the target area. The gantry angle may relate to the orientation of a radiation source (e.g., a therapeutic radiation source) relative to the isocenter of a radiotherapy device. For example, the gantry angle may be the angle between the vertical direction and the beam axis of a radiation beam emitted from the radiation source. For example, the therapeutic radiation source may be detachably or fixedly coupled to the gantry. When the gantry rotates about a gantry rotation axis, the therapeutic radiation source may rotate with the gantry, thereby enabling imaging and / or treatment of a subject on the scanning couch from different gantry angles.

[0096] In some embodiments, the processing device 140 can obtain a planning image of the target object. The planning image can refer to an image used to formulate a target plan for the target object. The planning image can include information related to the treatment area of ​​the target object. For example, the planning image can show a tumor of the target object and tissues and / or organs near the tumor. The planning image can be a CT image, an MRI image, etc. In some embodiments, the planning image can be a historical medical image of the target object. In some embodiments, the planning image can be the latest medical image obtained before performing a radiotherapy operation on the target object. A user of the processing device 140 or the medical system 100 (e.g., a doctor) can determine the target plan based on the planning image of the target object.

[0097] In some embodiments, the target plan of the target object can be stored in a storage device (e.g., storage device 150) of the medical system 100 or an external storage device, and the processing device 140 can obtain the target plan from the storage device (e.g., storage device 150) of the medical system 100 or an external storage device.

[0098] In some embodiments, the processing device 140 may determine, based on the target plan for the target subject, a characteristic related to the complexity of the target plan. The complexity of the target plan may reflect the difficulty of performing radiotherapy on the target subject based on the radiotherapy parameters involved in the target plan. In some embodiments, the complexity of the target plan is related to the values ​​of the radiotherapy parameters during the radiotherapy process based on the target plan. For example, the complexity of the target plan is related to the magnitude of the radiotherapy parameter values, the degree of change in the radiotherapy parameter values, the speed of change in the radiotherapy parameter values, or any combination thereof.

[0099] In some embodiments, the larger the radiotherapy parameter value in the target plan, the more complex the target plan. For example, the more sub-fields there are during radiotherapy, the more complex the target plan. In some embodiments, the greater the degree of change in the radiotherapy parameter value in the target plan, the more complex the target plan. For example, the greater the degree of change in the dose rate during radiotherapy, the more complex the target plan. In some embodiments, the faster the rate of change in the radiotherapy parameter value in the target plan, the more complex the target plan. For example, the faster the rate of change in the gantry angle during radiotherapy, the more complex the target plan.

[0100] In some embodiments, the features related to the complexity of the target plan may include the degree of irregularity of the radiation beam of the radiotherapy device, the product of the number of machine jumps corresponding to the field and the field area, the field perimeter, the small field ratio, the degree to which the field deviates from the center point of the medical device, the degree of change in the position of the collimator blades between adjacent sub-fields, etc., or any combination thereof. The small field ratio may refer to the ratio of the number of collimator blades whose field is less than a threshold value (e.g., 1 cm, 2 cm) during field opening to the total number of collimator blades involved in field opening. For example, assuming that there are 10 pairs of collimator blades involved in field opening, of which 2 pairs of collimator blades have a field opening less than a threshold value (e.g., 2 cm), the small field ratio is one-fifth. In some embodiments, the higher the degree of irregularity of the radiation beam of the radiotherapy device, the greater the product of the number of machine jumps corresponding to the field and the field area, the smaller the field perimeter, the higher the small field ratio, the higher the degree to which the field deviates from the center point of the medical device, etc., and / or the greater the degree of change in the position of the collimator blades between adjacent sub-fields, the higher the complexity of the target plan.

[0101] In some embodiments, the processing device 140 may determine a target flux map based on a target plan for the target subject. In some embodiments, the target flux map may be used to characterize the flux distribution of a beam passing through a beam limiter. For example, the target flux map may represent the expected radiation intensity distribution of a beam planned to be delivered to a target region of the subject during radiotherapy. In some embodiments, based on the target plan for the target subject, the radiotherapy device may perform radiotherapy on the target subject on the scanning couch from multiple gantry angles for each radiation beam. For each of the multiple gantry angles, the processing device 140 may determine multiple first sub-flux maps corresponding to the multiple sub-fields based on the radiation intensity per unit time and the radiation time corresponding to the multiple sub-fields at the gantry angle. The processing device 140 may determine a second sub-flux map corresponding to the gantry angle based on the multiple first sub-flux maps. For example, the processing device 140 may determine the second sub-flux map by weighted fusion of the multiple first sub-flux maps. Furthermore, the processing device 140 may determine a target flux map corresponding to the radiation beam based on the multiple second sub-flux maps corresponding to the multiple gantry angles. For example, the processing device 140 may determine a target flux map corresponding to the radiation beam by weighted fusion of the plurality of second sub-flux maps. In some embodiments, since the target flux map may be determined based on radiotherapy parameters using a dose model, the target flux map may also reflect, to a certain extent, information related to radiotherapy parameters (e.g., parameters related to a radiation field or sub-field) and parameter information of the dose model.

[0102] In 530 , the processing device 140 (eg, the determination module 430 ) may determine whether the status information of the medical device satisfies a preset condition.

[0103] In some embodiments, the processing device 140 may determine whether beam information (e.g., beam flatness, beam symmetry, beam linearity, beam output consistency, etc.) at one or more dose rates meets corresponding preset conditions (e.g., a flatness threshold, a symmetry threshold, a linearity threshold, an output consistency threshold). If it is determined that the beam information (e.g., beam flatness, beam symmetry, beam linearity, beam output consistency, etc.) at one or more dose rates meets corresponding preset conditions (e.g., above a flatness threshold, above a symmetry threshold, above a linearity threshold, above an output consistency threshold), the processing device 140 may determine that the beam information meets the preset conditions and the medical device status information check passes.

[0104] In some embodiments, processing device 140 may determine whether the in-place accuracy information of at least one component of the medical device is greater than a corresponding accuracy threshold. If it is determined that the in-place accuracy information of at least one component is greater than the corresponding accuracy threshold, processing device 140 may determine that the in-place accuracy information of the at least one component meets a preset condition, and the status information check of the medical device has passed.

[0105] In some embodiments, processing device 140 may determine whether an operational error of at least one component of the medical device is less than a corresponding error threshold. If it is determined that the operational error of at least one component is less than the corresponding error threshold, processing device 140 may determine that the operational error of the at least one component meets a preset condition and that the medical device status information check has passed.

[0106] In some embodiments, the processing device 140 can determine whether the status information of the medical device meets a preset condition based on the execution result of the data set of the medical device (e.g., the difference between the actual test result and the predicted test result). For example, the processing device 140 can determine whether the execution result of the data set passes based on the difference between the test image and the simulation image captured by an imaging device (e.g., an electronic portal imaging device) and a first difference threshold, thereby determining whether the status information of the medical device meets the preset condition. In some embodiments, the difference between the first image (e.g., the test image) and the second image (e.g., the simulation image) can be represented by the difference between the average grayscale value of the pixels or voxels of the first image and the average grayscale value of the pixels or voxels of the second image. If it is determined that the difference between the test image and the simulation image is greater than the first difference threshold, the processing device 140 can determine that the execution result of the data set fails and determines that the status information of the medical device does not meet the preset condition. If it is determined that the difference between the test image and the simulation image is less than the first difference threshold, the processing device 140 can determine that the execution result of the data set passes and determines that the status information of the medical device meets the preset condition. For another example, the processing device 140 may determine whether the dataset execution result passes based on the difference between the measured dose distribution and the planned dose distribution, thereby determining whether the status information of the medical device satisfies a preset condition. In some embodiments, the processing device 140 may determine whether the status information of the medical device satisfies a preset condition based on a pass rate of the dataset execution result of the medical device. If the pass rate of the dataset execution result of the medical device is determined to be greater than a pass rate threshold, the processing device 140 may determine that the status information of the medical device satisfies the preset condition.

[0107] In some embodiments, the same data set (e.g., a second data set, a third data set) can be executed multiple times using a medical device to obtain multiple execution results (e.g., multiple differences between actual test results and predicted test results). The processing device 140 can determine whether the status information of the medical device meets the preset conditions based on the multiple test results. For example, during the factory inspection of the medical device, the first execution result (e.g., EPID image, dose distribution) can be determined by executing the third data set using the medical device. After the medical device has been used for a period of time, the third data set can be executed again using the medical device to determine the second execution result (e.g., EPID image, dose distribution). The processing device 140 can determine whether the difference between the first execution result and the second execution result is less than a second difference threshold. If it is determined that the difference between the first execution result and the second execution result is less than the second difference threshold, the processing device 140 can determine that the status information of the medical device meets the preset conditions. In some embodiments, for the second and third data sets, the processing device 140 can determine whether the status information of the medical device meets the preset conditions based on the degree of change in the pass rate of the data set execution results of the medical device (e.g., the difference between the current pass rate of the data set execution results and the historical pass rate of the data set execution results). If it is determined that the degree of change in the pass rate of the data set execution result of the medical device is lower than the pass rate change degree threshold, the processing device 140 may determine that the status information of the medical device meets the preset condition.

[0108] In some embodiments, the preset conditions (e.g., a flatness threshold, a symmetry threshold, a linearity threshold, an output consistency threshold, an accuracy threshold, an error threshold, a first difference threshold, a second difference threshold, a pass rate threshold, a pass rate variation threshold) can be system recommended values ​​or system default values. In some embodiments, the preset conditions can be manually set by a user of the medical system 100 (e.g., a doctor, a technician) or automatically set by one or more components of the medical system 100 (e.g., the processing device 140) according to different circumstances.

[0109] In some embodiments, because each type of medical device status information changes with varying degrees of ease, the frequency of inspection required for each type of status information also varies accordingly. For some medical device parameters that are easily changed (e.g., accelerator output, multi-leaf collimator alignment accuracy), the inspection frequency needs to be higher (e.g., daily, weekly) to ensure the accuracy of the radiotherapy process. For some medical device parameters that are less likely to change (e.g., gantry rotation alignment accuracy, isocenter accuracy, and scanner bed alignment accuracy), the inspection frequency can be relatively low (e.g., monthly, annual) to save on the total time required for quality assurance testing and improve the overall efficiency of quality assurance testing.

[0110] In some embodiments, because the accuracy of a medical device executing a radiotherapy plan that is more common in clinical practice can be reflected by executing the first dataset, the frequency of status checks on the medical device based on the first dataset can be higher (e.g., daily checks, weekly checks). Because the accuracy of a medical device executing a radiotherapy plan that is less common in clinical practice can be reflected by executing the second and / or third datasets, the frequency of status checks on the medical device based on the second and / or third datasets can be lower (e.g., monthly checks, annual checks). In some embodiments, the user can set the frequency of checking each type of status information on the medical device based on experience and actual needs.

[0111] In 540 , in response to determining that the status information of the medical device satisfies the preset condition, the processing device 140 (eg, the determination module 430 ) may determine a prediction result based on the status information of the medical device and the target plan.

[0112] In some embodiments, the processing device 140 can determine the prediction results based on the status information and target plan of the medical device according to the quality assurance model. The quality assurance model can refer to a machine learning model or algorithm used to determine the prediction results based on the status information of the medical device and the target plan of the target object. The prediction results may include predicted EPID images, gamma pass rates, dose distributions (e.g., dose volume histograms), reasons for failure of quality assurance tests, target adjustment methods, etc. or any combination thereof. In some embodiments, the prediction results may also include dose measurement results of a dose measurement device. The dose measurement device may include a dose film, a water tank, a rotating irradiation dose meter (Arccheck), a multi-sequence planar dosimeter (MapCheck), etc.

[0113] In some embodiments, the processing device 140 may also acquire medical images of the target subject (e.g., scan data) and use the medical images of the target subject as one of the inputs to the quality assurance model. In this case, the quality assurance model may output a three-dimensional dose distribution within the target subject. For example, the processing device 140 may input the status information of the medical device, the target plan of the target subject, and the medical images of the target subject into the dose assurance model. The dose assurance model may output a three-dimensional dose distribution within the target subject (e.g., a dose-volume histogram). The medical images of the target subject may include CT images, PET images, MRI images, etc., or any combination thereof. In some embodiments, the medical images of the target subject may be medical images acquired during historical imaging or treatment of the target subject. In some embodiments, the medical images of the target subject may be the most recent medical images acquired before the current radiotherapy operation (e.g., a few days or hours ago) on the target subject, and may reflect the current state of the target subject. For example, the medical images of the target subject may be planning images used to determine the target plan. For another example, the medical images of the target subject may be positioning images of the target subject before each treatment fraction. According to some embodiments of the present specification, using a medical image of a target object as one of the inputs of a quality assurance model can improve the accuracy and rationality of the prediction results (e.g., dose distribution) output by the quality assurance model, while making the prediction results more clinically meaningful.

[0114] In some embodiments, the quality assurance model may include a first model and a second model. The first model is configured to determine a dose distribution (e.g., a two-dimensional dose distribution, a three-dimensional dose distribution), a predicted EPID image, and / or a gamma transmission rate based on the medical device's status information, a target plan for a target subject, parameter information of a dose model, a medical image of the target subject, or any combination thereof. In some embodiments, when the output of the first model is a predicted EPID image, a target flux map of the target subject is required as one of the inputs to the first model. For example, the medical device's status information and the target flux map may be input to the first model, and the first model may output a predicted EPID image. For another example, the medical device's status information, the target flux map, and features related to the complexity of the target plan may be input to the first model, and the first model may output a predicted EPID image. For another example, the medical device's status information, the target flux map, and parameter information of a dose model may be input to the first model, and the first model may output a predicted EPID image. For another example, the medical device's status information, the target flux map, features related to the complexity of the target plan, and parameter information of a dose model may be input to the first model, and the first model may output a predicted EPID image. For another example, the medical device's status information, the target flux map, features related to the complexity of the target plan, and parameter information of a dose model may be input to the first model, and the first model may output a predicted EPID image. For another example, the target flux map or at least one of the features related to the complexity of the target plan, and the status information of the medical device can be input into the first model, and the first model can output the gamma pass rate. For another example, the target flux map or at least one of the features related to the complexity of the target plan, the status information of the medical device and the parameter information of the dose model can be input into the first model, and the first model can output the gamma pass rate. For another example, the status information of the medical device, the target plan of the target object and the medical image of the target object can be input into the first model, and the first model can output the dose distribution. In some embodiments, when the target flux map is used as the input of the quality assurance model, since the target flux map can reflect the information related to the radiotherapy parameters (for example, parameters related to the field or sub-field) and the parameter information of the dose model to a certain extent, the information related to the radiotherapy parameters and the parameter information of the dose model may not need to be used as the input of the quality assurance model.

[0115] The second model is used to determine the reason for failure of the quality assurance test and / or the target adjustment method based on the status information of the medical device, the target plan of the target object, the parameter information of the dose model, the predicted EPID image, the medical image of the target object, etc., or any combination thereof. For example, the predicted EPID image (and / or calculated EPID image), the status information of the medical device, and the target plan of the target object (e.g., the target flux map and / or features related to the complexity of the target plan) can be input into the second model, and the second model can output the reason for failure of the quality assurance test. For another example, the predicted EPID image (and / or calculated EPID image), the status information of the medical device, the target plan of the target object (e.g., the target flux map and / or features related to the complexity of the target plan), and the parameter information of the dose model can be input into the second model, and the second model can output the reason for failure of the quality assurance test. More description of determining the prediction results can be found elsewhere in this specification (e.g., Figure 6 and its related description).

[0116] In some embodiments, the quality assurance model can be a machine learning model. For example, the quality assurance model can be a convolutional neural network model (CNN), a recurrent neural network model (RNN), a deep neural network model (DNN), a fully convolutional neural network (FCN) model, a generative adversarial network (GAN) model, etc. In some embodiments, a training algorithm can be used to generate a quality assurance model by training an initial model based on multiple sets of training samples. Exemplary training algorithms may include a gradient descent algorithm, a Newton algorithm, a quasi-Newton algorithm, a conjugate gradient algorithm, a generative adversarial learning algorithm, etc. More descriptions of the training process of the quality assurance model can be found elsewhere in this specification (e.g., Figure 7 and its related description).

[0117] In some embodiments, the processing device 140 can determine a prediction result using an analytical method based on the medical device's status information and target plan. For example, one or more of the following information, such as the medical device's status information, the target plan, a medical image of the target subject, and parameter information of the dose model, can serve as independent variables in an empirical dose calculation formula, and the prediction result can serve as a dependent variable in the empirical dose calculation formula. The empirical dose calculation formula can be determined based on historical radiotherapy data and / or simulation experimental data.

[0118] In 550 , the processing device 140 (eg, the determination module 430 ) may determine whether the quality assurance test passes based on the prediction result.

[0119] For example, the processing device 140 may determine whether the quality assurance test has passed based on the gamma pass rate and a first threshold. If it is determined that the gamma pass rate is less than the first threshold, the processing device 140 may determine that the quality assurance test has failed. If it is determined that the gamma pass rate is greater than or equal to the first threshold, the processing device 140 may determine that the quality assurance test has passed. For another example, the processing device 140 may determine whether the quality assurance test has passed based on the difference between the predicted image and the calculated image and a second threshold. If it is determined that the difference between the predicted image and the calculated image is greater than the second threshold, the processing device 140 may determine that the quality assurance test has failed. If it is determined that the difference between the predicted image and the calculated image is less than or equal to the second threshold, the processing device 140 may determine that the quality assurance test has passed.

[0120] In some embodiments, the target object may include multiple regions of interest, and the prediction result may include dose distributions corresponding to the multiple regions of interest (e.g., dose volume histograms (DVHs)). The regions of interest of the target object may include the planned target volume (PTV), organs at risk (OARs), other organs or tissues, etc. of the target object, or any combination thereof. The processing device 140 may obtain a weight value corresponding to each region of interest in the multiple regions of interest. The weight value corresponding to the region of interest may reflect the importance of the dose difference (or gamma pass rate) between the planned dose distribution and the predicted dose distribution of the region of interest in determining whether the quality assurance test has passed. For example, the higher the weight value corresponding to the region of interest, the higher the importance of the dose difference (or gamma pass rate) between the planned dose distribution and the predicted dose distribution of the region of interest in determining whether the quality assurance test has passed. In some embodiments, the weight values ​​corresponding to the organs at risk and the planned target volume are usually set higher to ensure that the tumor lesion area (i.e., the planned target volume) can receive the expected dose of radiation as much as possible during radiotherapy, and can minimize damage to surrounding normal tissues (i.e., organs at risk).

[0121] The processing device 140 can determine whether a quality assurance test has passed based on the weight value corresponding to each of the multiple regions of interest and the dose distribution corresponding to the multiple regions of interest. For example, for each of the multiple regions of interest, the processing device 140 can determine whether the product of the dose difference between the planned dose distribution and the predicted dose distribution of the region of interest and the weight value corresponding to the region of interest is greater than a corresponding dose difference threshold. In response to determining that the product of the dose difference between the planned dose distribution and the predicted dose distribution of the region of interest and the weight value corresponding to the region of interest is greater than the corresponding dose difference threshold, the processing device 140 can determine that the quality assurance test for the region of interest has failed. In some embodiments, when the quality assurance tests corresponding to the regions of interest with higher weights all pass, the processing device 140 can determine that the quality assurance test for the corresponding target plan has passed. In some embodiments, when the quality assurance tests for the regions of interest greater than a certain number of thresholds (e.g., 90%, 95%, 99%) pass, the processing device 140 can determine that the quality assurance test for the corresponding target plan has passed. In some embodiments, when the quality assurance tests for each region of interest pass, the processing device 140 can determine that the quality assurance test for the corresponding target plan has passed.

[0122] In some embodiments, the first threshold, the second threshold, the dose difference threshold and / or the quantity threshold can be manually set by a user of the medical system 100 (e.g., a doctor, a technician) or automatically set by one or more components of the medical system 100 (e.g., the processing device 140) according to different circumstances.

[0123] According to some embodiments of the present specification, by determining the predicted dose distribution (e.g., DVH) of each region of interest in the target object, the difference between the predicted dose distribution and the planned dose distribution of each region of interest in the target object can be comprehensively and intuitively displayed, making the quality assurance test results more clinically meaningful, and reasonable weight values ​​and dose difference thresholds can be set according to the characteristics of each region of interest (e.g., whether it is an eye tissue such as the lens, a planned target area, or an organ at risk), thereby improving the rationality and accuracy of the quality assurance test.

[0124] In response to determining that the quality assurance test passes, process 500 proceeds to step 560 where the processing device 140 (eg, control module 440 ) may perform a target operation on the target object.

[0125] In some embodiments, the target operation may include: the processing device 140 may control at least one component of the medical device to perform a radiotherapy operation on the target object based on the radiotherapy parameter values ​​in the target plan for the target object. In some embodiments, the target operation may include: the user of the medical system 100 (e.g., a doctor, a technician) adjusting or confirming the target plan based on the prediction result, and the processing device 140 may control at least one component of the medical device to perform the radiotherapy operation on the target object based on the adjusted or confirmed radiotherapy parameter values ​​in the target plan. In some embodiments, the target operation may include: the user of the medical system 100 (e.g., a doctor, a technician) performing quality assurance testing by using the medical device to execute the radiotherapy parameter values ​​in the target plan based on the prediction result.

[0126] In response to determining that the quality assurance test has failed, process 500 proceeds to step 570 where the processing device 140 (eg, determination module 430 ) may adjust the target plan or perform quality assurance testing on the target object.

[0127] In some embodiments, in response to determining that a quality assurance test has failed, the complexity of the target plan may be appropriately adjusted (e.g., reduced). In some embodiments, in response to determining that a quality assurance test has failed, the processing device 140 may determine the reason for the quality assurance test failure based on the second model, based on the medical device's status information, the target plan for the target subject, parameter information of the dose model, and the predicted image. The processing device 140 may adjust the corresponding parameter values ​​based on the reason for the quality assurance test failure. The processing device 140 may determine an updated prediction result based on the first model based on the adjusted parameter values. The processing device 140 may determine a target adjustment method based on the updated prediction result. For example, the processing device 140 may determine whether the quality assurance test has passed based on the updated prediction result. In response to determining that the quality assurance test has passed, the processing device 140 may determine the adjusted parameter values ​​as the target adjustment method. The processing device 140 may control the medical device to execute the updated target plan on the target subject. The updated target plan may be determined based on the adjusted parameter values. More description of how to determine the target adjustment can be found elsewhere in this specification (eg, steps 630 to 650 in FIG. 6 ).

[0128] In some embodiments, in response to determining that a quality assurance test has failed, a quality assurance test may be performed on the target subject based on the target plan. For example, the quality assurance test may be performed by using a medical device to execute the radiotherapy parameter values ​​in the target plan. In response to determining that the quality assurance test has passed, the processing device 140 may control at least one component of the medical device to perform radiotherapy on the target subject based on the radiotherapy parameter values ​​in the target plan for the target subject. In response to determining that the quality assurance test has failed, the processing device 140 may adjust the target plan until it is determined that the quality assurance test has passed.

[0129] It should be noted that the above description of this specification is provided for illustrative purposes only and is not intended to limit the scope of this specification. A person of ordinary skill in the art may make various changes and modifications based on the description of this specification. However, such changes and modifications do not depart from the scope of this specification. In some embodiments, steps 510-520 may be combined into a single step. In some embodiments, a step of obtaining parameter information of the dose model may be added to process 500. Accordingly, in step 540, the processing device 140 may determine a prediction result based on the medical device status information, the target plan, and the parameter information of the dose model. In some embodiments, a step of obtaining a medical image of the target object may be added to process 500. Accordingly, in step 540, the processing device 140 may determine a prediction result based on the medical device status information, the target plan, the parameter information of the dose model, and the medical image of the target object. In some embodiments, step 530 may be omitted, and the processing device 140 may directly determine the prediction result based on the medical device status information obtained in step 510 and the target plan obtained in step 520, without determining whether the medical device status information meets preset conditions. In some embodiments, the order of the steps in process 500 may be changed. For example, step 530 may be performed before step 520 .

[0130] In some embodiments, steps 510 and 530 may be performed periodically (e.g., daily, weekly, monthly, or annually) to check whether the status information of the medical device meets the requirements. Before performing radiotherapy on the target subject, steps 520, 540, and 550 are performed to determine the prediction results and, based on the prediction results, determine whether the quality assurance test has passed. In some embodiments, steps 510-550 are performed before performing radiotherapy on the target subject.

[0131] FIG6 is a flowchart of an exemplary process of a quality assurance method according to some embodiments of the present specification. In some embodiments, at least a portion of process 600 can be performed by a processing device 140 (e.g., implemented in the computing device 200 shown in FIG2 ). For example, process 600 can be stored in a storage device (e.g., storage device 150, memory 220, memory 390) in the form of instructions (e.g., application programs) and called and / or executed by a processing device 140 (e.g., the processor 210 shown in FIG2 , the CPU 340 shown in FIG3 , or one or more modules in the processing device 140 shown in FIG4 ). The operations of the process shown below are for illustrative purposes only. In some embodiments, process 600 can be completed using one or more additional operations not described and / or without one or more operations discussed. In addition, the order of the operations of process 600 shown in FIG6 and described below is not intended to be limiting.

[0132] In 610 , the processing device 140 (eg, the determination module 430 ) may determine at least one of a predicted image, a gamma pass rate, or a dose distribution by inputting the state information of the medical device and the target plan of the target object into the first model.

[0133] In some embodiments, the processing device 140 may determine a first eigenvector based on the status information of the medical device. For example, the processing device 140 may use the systematic error and random error of at least one component of the medical device as eigenvalues ​​in the first eigenvector. For another example, the processing device 140 may use beam and dose information (e.g., beam flatness, beam symmetry, beam linearity, etc.), positioning accuracy information of at least one component of the medical device (e.g., gantry, scanning table, radiotherapy radiation source, beam limiter), operational error of at least one component of the medical device, and data set execution results of the medical device as eigenvalues ​​in the first eigenvector. The processing device 140 may determine a second eigenvector based on the target plan of the target object. For example, the processing device 140 may use features related to the complexity of the target plan and the target flux map as eigenvalues ​​in the second eigenvector. The processing device 140 may input the first and second eigenvectors into a first model, and the first model may output at least one of a predicted image, a gamma pass rate, or a dose distribution. For example, the first model can output a predicted image (e.g., a predicted EPID image), and the processing device 140 can determine the gamma pass rate through gamma analysis based on the predicted image and the calculated image (e.g., the calculated EPID image). The calculated image (e.g., the calculated EPID image) can be an EPID image determined by the medical system 100 (e.g., the treatment planning system (TPS) in the medical system 100) based on the target plan of the target object. For another example, the first model can directly output the gamma pass rate. In some embodiments, the processing device 140 can input the first eigenvector and the second eigenvector into the first model, and the first model can output the dose distribution in the target object. For example, the first model can output a dose volume histogram (DVH). In some embodiments, the processing device 140 can determine one or more eigenvectors based on the parameter information of the medical image of the target object and / or the dose model, and use the one or more eigenvectors as input information of the first model.

[0134] In some embodiments, the processing device 140 may input a first target flux map into a first feature extraction model (e.g., a two-dimensional feature extraction model). In this embodiment, the first target flux map may be determined based on a medical device without systematic errors and / or random errors. The first feature extraction model may determine a two-dimensional feature vector corresponding to the first target flux map. The processing device 140 may input the systematic errors and / or random errors of at least one component of the medical device into a second feature extraction model (e.g., a one-dimensional feature extraction model). The second feature extraction model may determine a one-dimensional feature vector corresponding to the systematic errors and / or random errors. The processing device 140 may input the one-dimensional feature vector corresponding to the systematic errors and / or random errors into a dimensionality conversion model. The dimensionality conversion model may convert the one-dimensional feature vector corresponding to the systematic errors and / or random errors into a two-dimensional feature vector corresponding to the systematic errors and / or random errors. Further, the processing device 140 may input the two-dimensional feature vector corresponding to the first target flux map and the two-dimensional feature vector corresponding to the systematic errors and / or random errors into the first model, and the first model may output a predicted image. The first feature extraction model and the second feature extraction model may refer to machine learning models or algorithms for converting the input information of the model into feature vectors. The dimension conversion model may refer to a machine learning model or algorithm for converting data dimensions (e.g., conversion between one-dimensional feature vectors, two-dimensional feature vectors, and three-dimensional feature vectors).

[0135] In some embodiments, the processing device 140 may input the second target flux map into a first feature extraction model (e.g., a two-dimensional feature extraction model). In this embodiment, the second target flux map may be determined based on a medical device that has systematic and / or random errors. The first feature extraction model may determine a two-dimensional feature vector corresponding to the second target flux map. Further, the processing device 140 may input the two-dimensional feature vector corresponding to the second target flux map into the first model, and the first model may output a predicted image.

[0136] In some embodiments, the processing device 140 may input the first target flux map into a first feature extraction model (e.g., a two-dimensional feature extraction model). In this embodiment, the first target flux map may be determined based on a medical device without systematic errors and / or random errors. The first feature extraction model may determine a two-dimensional feature vector corresponding to the first target flux map. The processing device 140 may input the two-dimensional feature vector corresponding to the first target flux map into a dimensionality conversion model. The dimensionality conversion model may convert the two-dimensional feature vector corresponding to the first target flux map into a one-dimensional feature vector corresponding to the first target flux map. The processing device 140 may then input the systematic errors and / or random errors of at least one component of the medical device, features related to the complexity of the target plan, parameter information of the dose model, etc., or any combination thereof, into a second feature extraction model (e.g., a one-dimensional feature extraction model). The second feature extraction model may determine a one-dimensional feature vector corresponding to the systematic errors and / or random errors, a one-dimensional feature vector corresponding to the complexity of the target plan, a one-dimensional feature vector corresponding to the parameter information of the dose model, etc., or any combination thereof. Furthermore, the processing device 140 can input a one-dimensional characteristic vector corresponding to the first target flux map, a one-dimensional characteristic vector corresponding to the systematic error and / or random error, a one-dimensional characteristic vector corresponding to the complexity of the target plan, a one-dimensional characteristic vector corresponding to the parameter information of the dose model, etc., or any combination thereof, into the first model, and the first model can output the gamma pass rate.

[0137] In this specification, the model input data in step 610 (for example, the status information of the medical device, the target plan of the target object, and the parameter information of the dose model) may also be referred to as initial model input data (for example, the initial status information of the medical device, the initial target plan, and the parameter information of the initial dose model), and the prediction results determined in step 610 (for example, the predicted image, the gamma pass rate, and the dose distribution) may also be referred to as the initial prediction results (for example, the initial predicted image, the initial gamma pass rate, and the initial dose distribution).

[0138] In 620 , the processing device 140 (eg, the determination module 430 ) may determine whether the quality assurance test passes based on at least one of the predicted image, the gamma pass rate, or the dose distribution.

[0139] For example, the processing device 140 may determine whether the quality assurance test has passed based on the gamma pass rate. For another example, the processing device 140 may determine whether the quality assurance test has passed based on the difference between the predicted image and the calculated image. For another example, the processing device 140 may determine whether the quality assurance test has passed based on the difference between the planned dose distribution and the predicted dose distribution. Further description of determining whether a quality assurance test has passed can be found elsewhere in this specification (e.g., step 550 in FIG. 5 ).

[0140] In some embodiments, in response to the quality assurance test passing, the processing device 140 may control the medical device to perform radiation therapy on the target object based on the target plan.

[0141] In 630, in response to determining that the quality assurance test has failed, the processing device 140 (e.g., the determination module 430) can determine the reason why the quality assurance test has failed based on the status information of the medical device, the target plan of the target object, the parameter information of the dose model and the predicted image according to the second model.

[0142] In some embodiments, the processing device 140 may determine a first eigenvector based on status information of the medical device. The processing device 140 may determine a second eigenvector based on a target plan for the target object. The processing device 140 may determine a third eigenvector based on parameter information of the dose model. The processing device 140 may determine a fourth eigenvector based on the predicted image. The processing device 140 may determine the reason for failure of the quality assurance test based on the second model based on at least one of the first eigenvector, the second eigenvector, the third eigenvector, and the fourth eigenvector. For example, the processing device 140 may input the first eigenvector, the second eigenvector, the third eigenvector, and the fourth eigenvector into the second model, and the second model may output the reason for failure of the quality assurance test. For another example, the processing device 140 may input the first eigenvector, the second eigenvector, and the fourth eigenvector into the second model, and the second model may output the reason for failure of the quality assurance test.

[0143] For example, the processing device 140 may input a first target flux map into a first feature extraction model (e.g., a two-dimensional feature extraction model). In this embodiment, the first target flux map may be determined based on a medical device without systematic errors and / or random errors. The first feature extraction model may determine a two-dimensional feature vector corresponding to the first target flux map. The processing device 140 may input the two-dimensional feature vector corresponding to the first target flux map into a dimension conversion model. The dimension conversion model may convert the two-dimensional feature vector corresponding to the first target flux map into a one-dimensional feature vector corresponding to the first target flux map. Then, the processing device 140 may input the systematic errors and / or random errors of at least one component of the medical device, features related to the complexity of the target plan, and parameter information of the dose model into a second feature extraction model (e.g., a one-dimensional feature extraction model). The second feature extraction model may determine a one-dimensional feature vector corresponding to the systematic errors and / or random errors, a one-dimensional feature vector corresponding to the complexity of the target plan, and a one-dimensional feature vector corresponding to the parameter information of the dose model. Furthermore, the processing device 140 can input a one-dimensional feature vector corresponding to the first target flux map, a one-dimensional feature vector corresponding to the systematic error and / or random error, a one-dimensional feature vector corresponding to the complexity of the target plan, and a one-dimensional feature vector corresponding to the parameter information of the dose model into the second model, and the second model can output the reason for the failure of the quality assurance test.

[0144] The reasons for failure of the quality assurance test may include reasons related to the medical device, reasons related to the target plan of the target object, reasons related to the parameters of the dose model, etc. or any combination thereof. Reasons related to the medical device may include large operating errors of the medical device (for example, large systematic errors and random errors of the multi-leaf collimator). Reasons related to the target plan of the target object may include high complexity of the target plan (for example, large or small field area, large degree of deviation of the field from the center point of the medical device). Reasons related to the parameters of the dose model may include inappropriate parameters of the dose model, etc.

[0145] In some embodiments, the second model may output multiple reasons for failing the quality assurance test and prioritize the multiple reasons for failing the quality assurance test. For example, the multiple reasons may be prioritized in descending order of likelihood of causing the quality assurance test to fail.

[0146] In 640 , the processing device 140 (eg, the determination module 430 ) may adjust corresponding parameter values ​​based on the reason why the quality assurance test failed.

[0147] For example, if it is determined that the reason why the quality assurance test failed is related to the medical device. The processing device 140 or the user (e.g., a physicist or technician) can adjust the operating parameters of at least one component of the medical device (e.g., improve the operating accuracy of the component). For another example, if it is determined that the reason why the quality assurance test failed is related to the target plan. The processing device 140 or the user (e.g., a doctor) can adjust the value of the radiotherapy parameter in the target plan (e.g., increase or decrease the value of the radiotherapy parameter) to reduce the complexity of the target plan. For another example, if it is determined that the reason why the quality assurance test failed is related to the parameters of the dose model. The processing device 140 or the user (e.g., a physicist or technician) can adjust the value of the parameter in the dose model (e.g., use another type of dose model, increase or decrease the value of the parameter in the dose model).

[0148] In some embodiments, when there are multiple reasons for the quality assurance test failing, the processing device 140 or the user (e.g., a doctor) can adjust the parameter values ​​corresponding to each reason for the quality assurance test failing separately, or adjust the parameter values ​​corresponding to at least two of the multiple reasons for the quality assurance test failing at the same time.

[0149] In 650 , the processing device 140 (eg, the determination module 430 ) may determine an updated prediction result according to the first model based on the adjusted parameter values.

[0150] In some embodiments, the processing device 140 may input the adjusted parameter values ​​and other initial model input data (i.e., other unadjusted initial model input data) into the first model, and the first model may output an updated predicted image, an updated dose distribution, and / or an updated gamma pass rate. For example, if it is determined that the reason for failing the quality assurance test is related to the target plan, after adjusting the values ​​of the radiotherapy parameters in the target plan, the processing device 140 may input the adjusted target plan and the initial state information of the medical device into the first model, and the first model may output an updated predicted image, an updated dose distribution, and / or an updated gamma pass rate. For another example, if it is determined that the reason for failing the quality assurance test is related to the medical device, after improving the operating accuracy of the components of the medical device, the processing device 140 may input the adjusted state information of the medical device and the initial target plan into the first model, and the first model may output an updated predicted image, an updated dose distribution, and / or an updated gamma pass rate.

[0151] In 660 , the processing device 140 (eg, the determination module 430 ) may determine a target adjustment method based on the updated prediction result.

[0152] In some embodiments, if the updated prediction result is better than the initial prediction result (for example, the updated gamma pass rate is higher than the initial gamma pass rate determined in step 610), the processing device 140 can determine the current parameter adjustment method (i.e., the parameter adjustment method in step 640) as the target adjustment method. For example, assuming that the reason why the quality assurance test failed is related to the target plan, the processing device 140 can reduce the field area in the target plan, and input the reduced field area and other initial model input data (for example, the initial state information of the medical device, the parameter information of the initial dose model) into the first model, and the first model can output the updated gamma pass rate. If the updated gamma pass rate is higher than the initial gamma pass rate, the processing device 140 can determine that the target adjustment method is to reduce the field area. In some embodiments, if the updated gamma pass rate is lower than the initial gamma pass rate, the processing device 140 can determine that the target adjustment method is to increase the field area.

[0153] In some embodiments, if the initial prediction result is better than the updated prediction result (for example, the updated gamma pass rate is lower than the initial gamma pass rate determined in step 610), the processing device 140 may readjust other parameter values ​​based on the reason for the quality assurance test failure (for example, other parameter values ​​corresponding to the reason for the quality assurance test failure or parameter values ​​corresponding to the reason for the quality assurance test failure of other priorities), as described in step 640. The processing device 140 may determine a second updated prediction result based on the adjusted parameter values ​​according to the first model, as described in step 650. The processing device 140 may determine a target adjustment method based on the second updated prediction result. Steps 640-660 may be repeatedly performed until the updated prediction result is better than the initial prediction result, and the processing device 140 may determine the current parameter adjustment method as the target adjustment method.

[0154] In some embodiments, the processing device 140 may determine whether the quality assurance test has passed based on the updated prediction result. In response to determining that the quality assurance test has passed, the processing device 140 may determine the current parameter adjustment method (i.e., the parameter adjustment method in step 640) as the target adjustment method. The processing device 140 may control the medical device to execute the updated target plan for the target subject, the updated target plan being determined based on the adjusted parameter values.

[0155] In some embodiments, the processing device 140 may issue a reminder to the user. The reminder may be in the form of a text prompt, a voice prompt, an image prompt, a video prompt, an indicator light prompt, or any combination thereof. In some embodiments, the reminder may include the quality assurance test result (e.g., pass or fail). In some embodiments, if it is determined that the quality assurance test failed, the reminder may include the reason for the quality assurance test failure and / or how to adjust the target.

[0156] It should be noted that the above description of this specification is provided for illustrative purposes only and is not intended to limit the scope of this specification. For those skilled in the art, various changes and modifications can be made based on the description of this specification. However, these changes and modifications do not depart from the scope of this specification. In some embodiments, the first model and the second model can be used in conjunction. For example, the output layer of the first model can be connected to the input layer of the second model. The processing device 140 can input the status information of the medical device, the target plan of the target object, the medical image of the target object and the parameter information of the dose model into the input layer of the first model, and the predicted image, dose distribution and / or gamma pass rate output by the first model can be directly input into the input layer of the second model, and the second model can output the predicted image, dose distribution, gamma pass rate and the reason for failure of the quality assurance test.

[0157] Figure 7 is a flowchart of an exemplary process for joint training of the first model and the second model according to some embodiments of this specification. In some embodiments, at least a portion of process 700 can be performed by a processing device 140 (e.g., implemented in the computing device 200 shown in Figure 2). For example, process 600 can be stored in a storage device (e.g., storage device 150, memory 220, memory 390) in the form of instructions (e.g., application programs) and called and / or executed by a processing device 140 (e.g., the processor 210 shown in Figure 2, the CPU 340 shown in Figure 3, or one or more modules in the processing device 140 shown in Figure 4). The operations of the process shown below are for illustrative purposes only. In some embodiments, process 700 can be completed using one or more additional operations not described and / or without one or more operations discussed. In addition, the order of the operations of process 700 shown in Figure 7 and described below is not intended to be limiting.

[0158] In 710 , the processing device 140 (eg, the determination module 430 ) may obtain multiple sets of training samples.

[0159] Each set of training samples in the plurality of sets of training samples may include sample status information of the medical device, sample plans, sample parameter information of the dose model, reference images, and reference reasons for failure of quality assurance tests. The reference images and reference reasons for failure of quality assurance tests may serve as gold standards for model training. The sample status information of the medical device may include beam and dose information, positioning accuracy information of at least one component of the medical device, operational errors (e.g., systematic errors, random errors) of at least one component of the medical device, results of executing a data set using the medical device, etc., or any combination thereof, as described in step 510. The sample plan may include radiotherapy parameters involved in the radiotherapy process, sample flux maps, and portions of features related to the complexity of the radiotherapy plan, as described in step 520. The sample parameter information of the dose model may include the type of dose model (e.g., the type of dose algorithm), parameters of the dose model (e.g., parameters in the dose algorithm), etc., as described in step 510.

[0160] In some embodiments, the training samples may include historical data related to historical scanning processes. For example, the training samples may include historical status information of medical equipment, historical radiotherapy plans, historical parameter information of dose models, historical images, and historical reasons for failure of quality assurance tests related to historical radiotherapy processes of multiple candidate subjects (e.g., patients).

[0161] In some embodiments, training samples may include simulated data. A user of the processing device 140 or medical system 100 may generate simulated data by adjusting historical data related to historical scans. For example, a sample plan may be determined by adjusting radiotherapy parameter values ​​in a historical radiotherapy plan to obtain a sample plan of a desired complexity (e.g., a higher complexity). Using simulated data as training samples increases the diversity and enrichment of training samples, thereby improving the accuracy and effectiveness of models trained based on the training samples.

[0162] In 720 , the processing device 140 (eg, the determination module 430 ) may generate a candidate image by inputting sample state information and sample plans of the medical devices in the training sample into the first initial sub-model.

[0163] In some embodiments, one or more parameter values ​​of the first initial model (and / or the second initial model) can be iteratively updated by performing multiple iterations. Exemplary parameters of the first initial model (and / or the second initial model) may include the size of the kernel of the layer, the total number of layers, the number of nodes in each layer, the connection weight between two connected nodes, the bias vector related to the node, etc., or any combination thereof. During each iteration, the processing device 140 can obtain a certain set of training samples from multiple sets of training samples to jointly train the first initial model (and / or the second initial model). The following is an example of an iteration process.

[0164] The first initial sub-model may refer to a machine learning model that needs to be trained to generate the first model. In some embodiments, the processing device 140 may input sample state information and sample plans of a medical device in a set of training samples from the plurality of sets of training samples into an input layer of the first initial sub-model. The first initial sub-model may determine a first predicted output (i.e., a candidate image) based on the sample state information and sample plans of the medical device.

[0165] In 730 , the processing device 140 (eg, the determination module 430 ) may generate candidate reasons for quality assurance test failure by inputting sample status information of the medical device, sample plan, sample parameter information of the dose model, and candidate images into the second initial sub-model.

[0166] The second initial sub-model may refer to a machine learning model that needs to be trained to generate the second model. In some embodiments, the sample status information of the medical device in the training sample, the sample plan, the sample parameter information of the dose model, and the candidate image output by the first initial sub-model may be input into the input layer of the second initial sub-model. The second initial sub-model may determine a second prediction output (i.e., a candidate reason for failing the quality assurance test) based on the sample status information of the medical device, the sample plan, the sample parameter information of the dose model, and the candidate image.

[0167] In 740 , the processing device 140 (eg, the determination module 430 ) may determine a value of a first loss function based on the candidate image and the reference image.

[0168] The first loss function can be used to evaluate the difference between the predicted output of the first initial sub-model (e.g., the candidate image) and the desired output (e.g., the reference image). In some embodiments, if the value of the first loss function is greater than a first threshold in the current iteration, the parameter values ​​of the first initial sub-model can be adjusted to reduce the value of the first loss function (i.e., reduce the difference between the candidate image and the reference image).

[0169] In 750 , the processing device 140 (eg, the determination module 430 ) may determine a value of a second loss function based on the candidate reasons for quality assurance test failure and the reference reasons for quality assurance test failure.

[0170] The second loss function can be used to evaluate the difference between the predicted output of the second initial sub-model (e.g., candidate causes for quality assurance test failure) and the expected output (e.g., reference causes for quality assurance test failure). In some embodiments, if the value of the second loss function is greater than a second threshold in the current iteration, the parameter values ​​of the second initial sub-model can be adjusted to reduce the value of the second loss function (i.e., reduce the difference between the candidate causes for quality assurance test failure and the reference causes).

[0171] In 760 , the processing device 140 (eg, the determination module 430 ) may determine a value of the target loss function based on the value of the first loss function and the value of the second loss function.

[0172] In some embodiments, the processing device 140 may obtain a first weight corresponding to the first loss function and a second weight corresponding to the second loss function. The first weight and the second weight respectively indicate the importance of the first loss function and the second loss function when jointly training the first initial sub-model and the second initial sub-model based on the target loss function. The processing device 140 may determine the value of the target loss function based on the value of the first loss function, the first weight corresponding to the first loss function, the value of the second loss function, and the second weight corresponding to the second loss function. For example, assuming that the value of the first loss function is A1, the value of the second loss function is A2, the first weight corresponding to the first loss function is W1, and the second weight corresponding to the second loss function is W2, then the value of the target function is A=A1×W1+A2×W2. In some embodiments, the first weight and the second weight may be manually set by a user of the medical system 100 (e.g., a doctor, a technician) or automatically set by one or more components of the medical system 100 (e.g., the processing device 140) according to different circumstances.

[0173] In 770 , the processing device 140 (eg, the determination module 430 ) may update at least one of the first initial sub-model and the second initial sub-model by optimizing a value of the target loss function to generate a first model and a second model.

[0174] In some embodiments, if the value of the target loss function is greater than a third threshold value in the current iteration, the processing device 140 may adjust the parameter values ​​of the first initial sub-model and / or the second initial sub-model to reduce the value of the target loss function. In some embodiments, the parameter values ​​of the first initial sub-model and / or the second initial sub-model may be updated by performing multiple iterations until the value of the target loss function is less than or equal to the third threshold value. If the value of the target loss function is less than or equal to the third threshold value in the current iteration, the processing device 140 may use the updated first initial sub-model and the updated second initial sub-model generated in the current iteration as the trained first model and the second model.

[0175] In some embodiments, if the value of the target loss function converges, the updated first initial sub-model and the updated second initial sub-model generated in the current iteration can be used as the trained first model and the second model. If the change in the value of the target loss function in two or more consecutive iterations is less than a fourth threshold, the value of the target loss function can be considered to have converged. In some embodiments, when the number of iterations performed during the training process is greater than a fifth threshold, the updated first initial sub-model and the updated second initial sub-model generated in the current iteration can be used as the trained first model and the second model.

[0176] In some embodiments, the first threshold, the second threshold, the third threshold, the fourth threshold and / or the fifth threshold can be manually set by a user of the medical system 100 (e.g., a doctor, a technician) or automatically set by one or more components of the medical system 100 (e.g., the processing device 140) according to different circumstances.

[0177] It should be noted that the above description of this specification is provided for illustrative purposes only and is not intended to limit the scope of this specification. For those skilled 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, in the joint training process 700 of the first initial sub-model and the second initial sub-model, the step of determining the value of the first loss function can be omitted, that is, the value of the target loss function can be the value of the second loss function, and the processing device 140 can jointly train the first initial sub-model and the second initial sub-model based on the value of the second loss function.

[0178] In some embodiments, the processing device 140 may separately train the first initial sub-model and the second initial sub-model to generate the first model and the second model, respectively. Taking the process of training the first model as an example, the processing device 140 may obtain multiple sets of first training samples. Each of the multiple sets of first training samples may include sample status information, sample plan, and reference prediction results for the medical device. The reference prediction results may include reference images and / or reference gamma pass rates. The processing device 140 may generate the first model by training the first initial sub-model based on the multiple sets of first training samples. The training process may include multiple rounds of iterations. In the first iteration, the processing device 140 may obtain the first initial sub-model. In other iterations, the processing device 140 may obtain the first updated sub-model generated in the previous round of iteration. The processing device 140 may input the sample status information and sample plan of the medical device in the first training samples into the input layer of the first initial sub-model (or the first updated sub-model), and input the reference prediction results in the first training samples into the output layer of the first initial sub-model (or the first updated sub-model). The first initial sub-model (or the first updated sub-model) can output a candidate prediction result, which includes a candidate image and / or a candidate gamma pass rate. The candidate prediction result can be compared with the reference prediction result based on the loss function. In some embodiments, multiple iterations can be performed to update the parameter value of the first initial sub-model until a termination condition is met. The termination condition can be related to the loss function or the number of iterations in the iterative process. For example, if the value of the loss function is less than a threshold, it can be determined that the termination condition is met. For another example, if the value of the loss function converges, it can be determined that the termination condition is met. For another example, when a specified number of iterations are performed during the training process, it can be determined that the termination condition is met. If it is determined that the iteration meets the termination condition, the first updated sub-model generated in the current iteration is determined as the first model. If it is determined that the iteration does not meet the termination condition, the parameter value of the first updated sub-model generated in the current iteration is adjusted until it is determined that the iteration meets the termination condition. Similarly, in the process of training the second model, the processing device 140 can obtain multiple groups of second training samples, each of the multiple groups of second training samples includes sample status information of the medical device, sample plan, sample parameter information of the dose model, sample images and reference prediction results, and the reference prediction results include the reasons for failure of the quality assurance test. The processing device 140 can generate a second model by training the second initial sub-model based on the multiple groups of second training samples. It should be noted that the above description of model training is provided for illustrative purposes only and is not intended to limit the scope of this specification. The first training samples and the second training samples may also include other information, or may not include the information mentioned above. For example, the first training samples may also include sample parameter information of the dose model and / or sample medical images.For another example, the first training sample and / or the second training sample may include features related to the complexity of the target plan and / or a target flux map. For another example, the second training sample may not include sample parameter information for the dose model. For another example, the dose assurance model may only include the first model, in which case only the first initial sub-model needs to be trained to generate the first model.

[0179] In some embodiments, the processing device 140 may first train the first initial sub-model and the second initial sub-model separately to generate the first candidate model and the second candidate model, respectively. Then, the processing device 140 may jointly train the first candidate model and the second candidate model to generate the first model and the second model. In some embodiments, the processing device 140 may first jointly train the first initial sub-model and the second initial sub-model to generate the first candidate model and the second candidate model, respectively. Then, the processing device 140 may separately train the first candidate model and the second candidate model to generate the first model and the second model.

[0180] In some embodiments, the quality assurance model (e.g., the first model, the second model) can be updated periodically or aperiodically. For example, the quality assurance model can be updated based on new training samples. In some embodiments, the generation, training, and / or updating of the quality assurance model can be performed online in response to a request for quality assurance testing. In some embodiments, the generation, training, and / or updating of the quality assurance model can be performed offline.

[0181] In some embodiments, the generation, training, and / or updating of a quality assurance model (e.g., a first model, a second model) can be performed on a first system, and quality assurance testing based on the quality assurance model can be performed on a second system. The first system and the second system can be the same or different. The first system and / or the second system can include the medical system 100. For example, the generation, training, and / or updating of the quality assurance model can be performed on a first processing device of the medical system 100, and the application of the quality assurance model can be performed on a second processing device of the medical system 100. The first processing device and the second processing device can be the same or different. The first processing device and / or the second processing device can include the processing device 140.

[0182] Compared with the prior art, the beneficial effects that may be brought about by the above embodiments of this specification include but are not limited to: (1) The operating state of the medical device will change as the use time of the medical device increases. The different operating states of the medical device have a greater impact on the quality assurance test results. Taking full account of the current operating error of the medical device during the quality assurance test can make the quality assurance test results more accurate and effective. (2) The data set execution results obtained by using the medical device to execute different types of data sets (for example, the first data set, the second data set, and the third data set) can reflect the accuracy of the medical device in executing different types of radiotherapy plans, so that the execution effect of the medical device in executing the radiotherapy plan can be determined without using a radiotherapy plan for a specific patient. (3) When performing dose calculation, in order to improve the efficiency of dose calculation, the medical device will be modeled. Using different dose models and different dose model parameters will have a greater impact on the quality assurance test results. Taking full account of the parameter information of the dose model during the quality assurance test can make the quality assurance test results more accurate and effective. (4) Based on one or more of the status information of the medical device, the target plan of the target object, and the parameter information of the dose model, the prediction results are determined by machine learning methods, which can save the time of quality assurance testing, improve the efficiency of quality assurance testing, and reduce the labor intensity of operators. (5) The reasons why the quality assurance test failed can be analyzed through the results of the quality assurance test, and further suggested adjustment methods can be given, thereby saving the operator's time and improving the efficiency of the radiotherapy process. (6) The quality assurance model can be updated and trained based on different training samples. For example, the quality assurance model can be updated and trained as radiotherapy technology develops and application scenarios change, thereby ensuring the rationality and accuracy of quality assurance tests performed based on the quality assurance model. It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects that may be produced may be any one or a combination of the above, or any other possible beneficial effects.

[0183] The basic concepts have been described above. It will be apparent to those skilled in the art after reading this application that the above disclosures are provided for illustrative purposes only and do not constitute limitations on this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and revisions to this specification. Such modifications, improvements, and revisions are suggested in this specification and remain within the spirit and scope of the exemplary embodiments of this specification.

[0184] This specification also uses specific terms to describe the embodiments of this specification. For example, "one embodiment," "an embodiment," and / or "some embodiments" refer to a feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that the mention of "one embodiment," "an embodiment," or "an alternative embodiment" two or more times in different locations in this specification does not necessarily refer to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of this specification may be appropriately combined.

[0185] In addition, it will be understood by those skilled in the art that various aspects of this specification 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 improvement thereof. Therefore, various aspects of this specification may be implemented entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may be referred to as a "unit," "module," or "system." In addition, various aspects of this specification may take the form of a computer program product embodied in one or more computer-readable media, wherein computer-readable program code is contained therein.

[0186] A computer-readable signal medium may include a propagated data signal embodying computer program code, for example, in baseband or as part of a carrier wave. Such propagated signals may take a variety of forms, including electromagnetic, optical, or any suitable combination. A computer-readable signal medium may be any computer-readable medium, other than a computer-readable storage medium, that can be coupled to an instruction execution system, apparatus, or device to communicate, propagate, or transfer a program for use. Program code on a computer-readable signal medium may be propagated via any suitable medium, including radio, cable, fiber optic cable, RF, or any combination of the foregoing.

[0187] The computer program code required for the operation of each part of this specification can be written in any one or more programming languages, including object-oriented programming languages ​​such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB.NET, Python, etc., conventional procedural programming languages ​​such as C programming language, Visual Basic, Fortran2103, Perl, COBOL2102, PHP, ABAP, dynamic programming languages ​​such as Python, Ruby and Groovy, or other programming languages. The program code can be run entirely on the user's computer, or run on the user's computer as an independent software package, or run partly on the user's computer and partly on a remote computer, or run entirely on a remote computer or server. In the latter case, the remote computer can be connected to the user's computer through any type of network (including a local area network (LAN) or a wide area network (WAN)), or a connection can be established with an external computer (for example, by using the network of a network service provider) or in a cloud computing environment or provided as a service, for example, software as a service (SaaS).

[0188] In addition, unless expressly stated in the claims, the order of the processing elements and sequences described in this specification, 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 specification. Although the above disclosure discusses some embodiments of the invention that are currently considered useful through various examples, it should be understood that such details are for illustrative purposes only, and the appended 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 spirit and scope of the embodiments of this specification. 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.

[0189] Similarly, it should be noted that, in order to simplify the presentation disclosed herein and thereby facilitate understanding of one or more embodiments of the invention, the foregoing descriptions of the embodiments of this specification sometimes combine various features into a single embodiment, figure, or description thereof. However, this approach should not be interpreted as reflecting an intention that the claimed object material to be scanned requires more features than expressly recited in each claim. Rather, the subject matter of the invention may include fewer features than the single embodiment described above.

[0190] In some embodiments, the numbers representing quantities or properties used to describe and claim certain embodiments of this specification should be understood as being modified in some cases by the terms "about," "approximately," or "substantially." For example, unless otherwise stated, "about," "approximately," or "substantially" can indicate a ±20% variation of the value it describes. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may vary depending on the desired characteristics of the individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical fields and parameters used to confirm the breadth of their range in some embodiments of this specification are approximate values, in specific embodiments, the settings of such numerical values ​​are as accurate as possible within the feasible range.

[0191] All patents, patent applications, patent application publications, and other materials (such as treatises, books, specifications, publications, records, things, and / or the like) referred to in this specification are hereby incorporated by reference in their entirety for all purposes, except any prosecution record related to such documents, any such documents that are inconsistent or conflicting with this document, or any such documents that limit the broad scope of the claims that may or may not be related to this document. For example, if there is any inconsistency or conflict between the description, definitions, and / or use of terms associated with any incorporated material and the terminology associated with this document, the description, definitions, and / or use of terms in this document will control.

[0192] Finally, it should be understood that the embodiments described in this specification are intended only to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly described and illustrated in this specification.

Claims

1. A quality assurance method, characterized in that: The method comprises: Obtain status information of medical equipment; Get the target plan of the target object; determining a prediction result based on the status information of the medical device and the target plan; and Based on the prediction results, it is determined whether the quality assurance test is passed.

2. The method according to claim 1, characterized in that The status information of the medical device includes at least one of beam information corresponding to multiple dose rates, positioning accuracy information of at least one component of the medical device at multiple positions, or operation error of at least one component of the medical device.

3. The method according to claim 1 or 2, characterized in that The obtaining of status information of the medical device includes: acquiring at least one data set, wherein the at least one data set is determined based on at least one of a candidate plan, a limit performance of at least one component of the medical device, or a calculated limit of a dose model; and By executing the at least one data set using the medical device, status information of the medical device is determined.

4. The method according to claim 3, characterized in that The determining of a prediction result based on the status information of the medical device and the target plan includes: determining whether the status information of the medical device satisfies a preset condition; and In response to determining that the status information of the medical device satisfies the preset condition, the prediction result is determined based on the status information of the medical device and the target plan.

5. The method according to claim 4, characterized in that The at least one data set includes at least one of a first data set, a second data set, or a third data set, and acquiring the at least one data set includes: determining the first data set based on at least one first candidate plan, wherein a value of a first candidate parameter in the at least one first candidate plan is within a first preset range; determining the second data set based on at least one second candidate plan, wherein a value of a second candidate parameter in the at least one second candidate plan is outside a second preset range; and The third data set is determined based on a limiting performance of at least one component of the medical device and / or a calculated limit of the dose model.

6. The method according to claim 5, characterized in that The at least one data set is the second data set or the third data set, and determining whether the status information of the medical device meets a preset condition includes: determining a plurality of test results by performing a plurality of tests based on the at least one data set; and Based on the multiple test results, it is determined whether the status information of the medical device meets a preset condition.

7. The method according to any one of claims 1 to 6, characterized in that The step of obtaining the target plan of the target object includes: Based on the target plan, characteristics and / or a target flux map are determined that are related to a complexity level of the target plan.

8. The method according to any one of claims 1 to 7, characterized in that The determining of a prediction result based on the status information of the medical device and the target plan includes: The prediction result is determined based on the status information of the medical device and the target plan of the target object according to a quality assurance model, wherein the quality assurance model is a machine learning model.

9. The method according to claim 8, characterized in that The quality assurance model includes a first model, and determining the prediction result according to the quality assurance model based on the status information of the medical device and the target plan of the target object includes: The prediction result is determined by inputting the status information of the medical device and the target plan of the target object into the first model, wherein the prediction result includes at least one of a predicted image, a gamma pass rate, or a dose distribution.

10. The method according to claim 9, characterized in that The quality assurance model includes a second model, and the method further includes: In response to determining that the quality assurance test has failed, a reason for the quality assurance test failure is determined according to the second model based on the status information of the medical device, the target plan of the target subject, and the prediction result.

11. The method according to claim 10, characterized in that The method further comprises: Adjust corresponding parameter values ​​based on the reasons why the quality assurance test failed; Determining an updated prediction result based on the first model based on the adjusted parameter values; determining whether the quality assurance test passes based on the updated prediction result; and In response to determining that the quality assurance test passes, the medical device is controlled to execute an updated target plan on the target object, wherein the updated target plan is determined based on the adjusted parameter value.

12. The method according to any one of claims 8 to 11, characterized in that The determining the prediction result based on the status information of the medical device and the target plan of the target object according to a quality assurance model includes: acquiring an image of the target object; and Based on the status information of the medical device, the target plan of the target object and the image of the target object, a three-dimensional dose distribution in the target object is determined according to the quality assurance model.

13. The method according to any one of claims 1 to 12, characterized in that The target object includes a plurality of regions of interest, the prediction results include dose distributions corresponding to the plurality of regions of interest, and determining whether a quality assurance test is passed based on the prediction results includes: Obtaining a weight value corresponding to each of the plurality of regions of interest; and Whether the quality assurance test is passed is determined based on the weight value corresponding to each region of interest in the multiple regions of interest and the dose distribution corresponding to the multiple regions of interest.

14. A quality assurance system, characterized in that The system comprises: at least one storage device storing an instruction set; and At least one processor in communication with the at least one storage device, wherein when executing the set of instructions, the at least one processor is configured to cause the system to perform the following operations: Obtain status information of medical equipment; Get the target plan of the target object; determining a prediction result based on the status information of the medical device and the target plan; and Based on the prediction results, it is determined whether the quality assurance test is passed.

15. The system according to claim 14, wherein: The status information of the medical device includes at least one of beam information corresponding to multiple dose rates, positioning accuracy information of at least one component of the medical device at multiple positions, or operation error of at least one component of the medical device.

16. The system according to claim 14 or 15, characterized in that The obtaining of status information of the medical device includes: acquiring at least one data set, wherein the at least one data set is determined based on at least one of a candidate plan, a limit performance of at least one component of the medical device, or a calculated limit of a dose model; and By executing the at least one data set using the medical device, status information of the medical device is determined.

17. The system according to claim 16, wherein: The determining of a prediction result based on the status information of the medical device and the target plan includes: determining whether the status information of the medical device satisfies a preset condition; and In response to determining that the status information of the medical device satisfies the preset condition, the prediction result is determined based on the status information of the medical device and the target plan.

18. The system according to claim 17, wherein: The at least one data set includes at least one of a first data set, a second data set, or a third data set, and acquiring the at least one data set includes: determining the first data set based on at least one first candidate plan, wherein a value of a first candidate parameter in the at least one first candidate plan is within a first preset range; determining the second data set based on at least one second candidate plan, wherein a value of a second candidate parameter in the at least one second candidate plan is outside a second preset range; and The third data set is determined based on a limiting performance of at least one component of the medical device and / or a calculated limit of the dose model.

19. The system according to claim 18, wherein: The at least one data set is the second data set or the third data set, and determining whether the status information of the medical device meets a preset condition includes: determining a plurality of test results by performing a plurality of tests based on the at least one data set; and Based on the multiple test results, it is determined whether the status information of the medical device meets a preset condition.

20. The system according to any one of claims 14 to 19, characterized in that The step of obtaining the target plan of the target object includes: Based on the target plan, characteristics and / or a target flux map are determined that are related to a complexity level of the target plan.

21. The system according to any one of claims 14 to 20, characterized in that The determining of a prediction result based on the status information of the medical device and the target plan includes: The prediction result is determined based on the status information of the medical device and the target plan of the target object according to a quality assurance model, wherein the quality assurance model is a machine learning model.

22. The system according to claim 21, wherein: The quality assurance model includes a first model, and determining the prediction result according to the quality assurance model based on the status information of the medical device and the target plan of the target object includes: The prediction result is determined by inputting the status information of the medical device and the target plan of the target object into the first model, wherein the prediction result includes at least one of a predicted image, a gamma pass rate, or a dose distribution.

23. The system according to claim 22, wherein: The quality assurance model includes a second model, and the at least one processor is configured to cause the system to further perform the following operations: In response to determining that the quality assurance test has failed, a reason for the quality assurance test failure is determined according to the second model based on the status information of the medical device, the target plan of the target subject, and the prediction result.

24. The system according to claim 23, wherein: The at least one processor is configured to cause the system to further perform the following operations: Adjust corresponding parameter values ​​based on the reasons why the quality assurance test failed; Determining an updated prediction result based on the first model based on the adjusted parameter values; determining whether the quality assurance test passes based on the updated prediction result; and In response to determining that the quality assurance test passes, the medical device is controlled to execute an updated target plan on the target object, wherein the updated target plan is determined based on the adjusted parameter value.

25. The system according to any one of claims 21 to 24, characterized in that The determining the prediction result based on the status information of the medical device and the target plan of the target object according to a quality assurance model includes: acquiring an image of the target object; and Based on the status information of the medical device, the target plan of the target object and the image of the target object, a three-dimensional dose distribution in the target object is determined according to the quality assurance model.

26. The system according to any one of claims 14 to 25, characterized in that The target object includes a plurality of regions of interest, the prediction results include dose distributions corresponding to the plurality of regions of interest, and determining whether a quality assurance test is passed based on the prediction results includes: Obtaining a weight value corresponding to each of the plurality of regions of interest; and Whether the quality assurance test is passed is determined based on the weight value corresponding to each region of interest in the multiple regions of interest and the dose distribution corresponding to the multiple regions of interest.

27. A non-transitory computer-readable medium comprising at least one set of instructions that, when executed by at least one processor of a computing device, causes the at least one processor to implement a method comprising: Obtain status information of medical equipment; Get the target plan of the target object; determining a prediction result based on the status information of the medical device and the target plan; as well as Based on the prediction results, it is determined whether the quality assurance test is passed.

28. A quality assurance system, characterized in that The system comprises: A first acquisition module is used to obtain status information of the medical device; A second acquisition module is used to acquire a target plan of a target object; and A determination module is configured to determine a prediction result based on the status information of the medical device and the target plan; and determine whether a quality assurance test is passed based on the prediction result.

29. A quality assurance device, characterized in that: The quality assurance device is configured to execute the quality assurance method according to any one of claims 1 to 13.

Images