A dose prediction method and system

Abstract

A dose prediction method and system, the method comprising: acquiring an original dose distribution of a radiotherapy object, original sign information and current sign information; obtaining a predicted Pareto optimization parameter through a dose prediction model according to the original dose distribution and the original sign information; and predicting a current dose distribution corresponding to the current sign information by using the dose prediction model according to the current sign information and the predicted Pareto optimization parameter.

Description

TECHNICAL FIELD

[0001] The present specification relates to the technical field of medical technology, and in particular, to a dose prediction method and system. BACKGROUND

[0002] Adaptive radiotherapy is a new radiotherapy mode that is currently popular. Before radiotherapy, a treatment plan is made for the patient, but the plan is based on the patient's physical information at that time. When the patient is treated, the physical information may change. Through the adaptive radiotherapy mode, the treatment plan can be quickly adjusted according to the current physical information of the patient to improve the treatment effect. The optimization problem of the radiotherapy plan is a multi-objective optimization problem, including the minimum maximum dose requirement of the target area, the maximum dose requirement of a certain protective organ, and the like. Generally, the user can set the weight of each target to generate a treatment plan. However, the weight has no clinical explanation, so the user can only give it according to experience. Alternatively, the user can use a multi-objective optimization system to find a plan that best meets the clinical requirements by adjusting the optimization parameters, but this process is very time-consuming.

[0003] Therefore, it is necessary to provide a more efficient dose prediction method and system. SUMMARY

[0004] One of the embodiments of the present specification provides a dose prediction method, which comprises: obtaining an original dose distribution, original physical information, and current physical information of a radiotherapy object; obtaining a predicted Pareto optimization parameter through a dose prediction model according to the original dose distribution and the original physical information; and predicting a current dose distribution corresponding to the current physical information using the dose prediction model according to the current physical information and the predicted Pareto optimization parameter.

[0005] One of the embodiments of the present specification provides a dose prediction system, which comprises: an acquisition module configured to obtain an original dose distribution, original physical information, and current physical information of a radiotherapy object; a first prediction module configured to obtain a predicted Pareto optimization parameter through a dose prediction model according to the original dose distribution and the original physical information; and a second prediction module configured to predict a current dose distribution corresponding to the current physical information using the dose prediction model according to the current physical information and the predicted Pareto optimization parameter.

[0006] One of the embodiments of the present specification provides a dose prediction device, which comprises: at least one storage medium storing computer instructions; and at least one processor executing the computer instructions to implement the dose prediction method as described above.

[0007] One of the embodiments of the present specification provides a computer readable storage medium, the storage medium stores computer instructions, when the computer reads the computer instructions, the computer executes the dose prediction method as described above.

[0008] The embodiments of the present specification directly locate the corresponding Pareto optimal parameters from the original dose distribution, thereby saving the process of generating respective Pareto planes and adjusting the Pareto optimization parameters for each case, making the automation degree of adaptive radiotherapy mode higher and the dose prediction effect better. BRIEF DESCRIPTION OF DRAWINGS

[0009] The present specification will be further described in the manner of exemplary embodiments, which will be described in detail through the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same numbers represent the same structures, in which:

[0010] Figure 1 is a schematic diagram of an application scenario of an exemplary dose prediction system according to some embodiments of the present specification;

[0011] Figure 2 is a module diagram of an exemplary dose prediction system according to some embodiments of the present specification;

[0012] Figure 3 is a flowchart of an exemplary dose prediction method according to some embodiments of the present specification;

[0013] Figure 4 is a schematic diagram of a first model according to some embodiments of the present specification;

[0014] Figure 5 is a schematic diagram of a second model according to some embodiments of the present specification. DETAILED DESCRIPTION

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the drawings required to be used in the embodiment description will be briefly introduced below. Obviously, the drawings in the following description are only some examples or embodiments of the present specification, and for those skilled in the art, the present specification can also be applied to other similar scenarios without creative labor. Unless it is obvious from the language environment or otherwise stated, the same reference numbers in the drawings represent the same structures or operations.

[0016] It should be understood that the "system", "device", "unit" and / or "module" used herein is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0017] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0018] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0019] Figure 1 This is a schematic diagram illustrating an application scenario of an exemplary dose prediction system according to some embodiments of this specification. In some embodiments, such as Figure 1 As shown, the application scenario 100 of the dose prediction system may include at least radiotherapy equipment 110, processing equipment 120, terminal equipment 130, storage device 140 and network 150.

[0020] In some embodiments, the radiotherapy device 110 may include a treatment apparatus. The treatment apparatus may include a linear accelerator, a cyclotron accelerator, a synchrotron, etc., configured to perform radiotherapy on the patient. The treatment apparatus may include accelerators corresponding to different types of particles, including, for example, photons, electrons, protons, or heavy ions.

[0021] In some embodiments, the radiotherapy device 110 may further include a scanning device capable of scanning a target object (treatment object) within a detection area or scanning area to obtain scan data of the target object. In some embodiments, the target object may include biological objects and / or non-biological objects. For example, the target object may be living or non-living organic and / or inorganic matter.

[0022] In some embodiments, the scanning device can include a single modality scanner and / or a multi-modality scanner. The single modality scanner can include, for example, an ultrasound scanner, an X-ray scanner, a computed tomography (CT) scanner, a magnetic resonance imaging (MRI) scanner, an ultrasonograph, a positron emission computed tomography (PET) scanner, an optical coherence tomography (OCT) scanner, an ultrasound (US) scanner, an intravascular ultrasound (IVUS) scanner, a near-infrared spectroscopy (NIRS) scanner, a far-infrared (FIR) scanner, or the like or any combination thereof. The multi-modality scanner can include, for example, an X-ray imaging-magnetic resonance imaging (X-ray-MRI) scanner, a positron emission tomography-X-ray imaging (PET-X-ray) scanner, a single photon emission computed tomography-magnetic resonance imaging (SPECT-MRI) scanner, a positron emission tomography-computed tomography (PET-CT) scanner, a digital subtraction angiography-magnetic resonance imaging (DSA-MRI) scanner, or the like or any combination thereof. The above descriptions of the scanning devices are for illustrative purposes only and are not intended to limit the scope of the present specification.

[0023] In some embodiments, the radiotherapy device 110 can include a treatment plan system (TPS), an image-guide radiotherapy (IGRT), or the like.

[0024] The processing device 120 can process data and / or information acquired from the radiotherapy device 110, the terminal device 130, the storage device 140, and / or other components of the application scenario 100 of the dose prediction system. For example, the processing device 120 can acquire and analyze the physical information and / or the dose distribution (e.g., the first nuclear sequence) of the radiotherapy subject from the terminal device 130, the storage device 140. For another example, the processing device 120 can control the radiotherapy device 110 to work based on the result of the analysis and processing.

[0025] In some embodiments, the processing device 120 can be a single server or a group of servers. The group of servers can be centralized or distributed. In some embodiments, the processing device 120 can be local or remote. For example, the processing device 120 can access information and / or data from the radiotherapy device 110, the terminal device 130, and / or the storage device 140 through the network 150. For another example, the processing device 120 can be directly connected to the radiotherapy device 110, the terminal device 130, and / or the storage device 140 to access information and / or data. In some embodiments, the processing device 120 can be implemented on a cloud platform. For example, the cloud platform can include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an inter-cloud, a multi-cloud, or the like or any combination thereof.

[0026] In some embodiments, the processing device 120 and the radiotherapy device 110 can be integrated. In some embodiments, the processing device 120 and the radiotherapy device 110 can be directly or indirectly connected to work together to implement the methods and / or functions described in this specification.

[0027] In some embodiments, the processing device 120 can include an input device and / or an output device. Through the input device and / or the output device, interaction with a user (e.g., setting a scan sequence, adjusting scan parameters, etc.) can be achieved. In some embodiments, the input device and / or the output device can include a display screen, a keyboard, a mouse, a microphone, etc., or any combination thereof.

[0028] The terminal device 130 can communicate and / or connect with the radiotherapy device 110, the processing device 120, and / or the storage device 140. In some embodiments, interaction with a user can be achieved through the terminal device 130. In some embodiments, the terminal device 130 can include a mobile device 131, a tablet 132, a laptop 133, etc., or any combination thereof. In some embodiments, the terminal device 130 (or all or part of its functions) can be integrated in the processing device 120.

[0029] The storage device 140 can store data, instructions, and / or any other information. In some embodiments, the storage device 140 can store data acquired from the radiotherapy device 110, the processing device 120, the terminal device 130, and / or the processing device 120 (e.g., dose distribution, physical information, Pareto optimization parameters, etc.). In some embodiments, the storage device 140 can store data and / or instructions used by the processing device 120 to perform or use to complete the exemplary methods described in this specification.

[0030] In some embodiments, the storage device 140 can include one or more storage components, each of which can be a separate device or part of another device. In some embodiments, the storage device 140 can include random access memory (RAM), read-only memory (ROM), mass storage, removable storage, volatile read-write memory, etc., or any combination thereof. In some embodiments, the storage device 140 can be implemented on a cloud platform. In some embodiments, the storage device 140 can be part of the radiotherapy device 110, the processing device 120, and / or the terminal device 130.

[0031] The network 150 can include any suitable network capable of facilitating the exchange of information and / or data. In some embodiments, at least one component of the application scenario 100 of the dose prediction system (e.g., the radiotherapy device 110, the processing device 120, the terminal device 130, the storage device 140) can exchange information and / or data with at least one other component of the application scenario 100 of the dose prediction system through the network 150. For example, the processing device 120 can obtain the target image of the target object from the radiotherapy device 110 through the network 150.

[0032] It should be noted that the above description of the application scenario 100 of the dose prediction system is provided only for the purpose of illustration and is not intended to limit the scope of the present specification. Various modifications or changes can be made according to the description of the present specification by those of ordinary skill in the art. For example, the application scenario 100 of the dose prediction system can implement similar or different functions on other devices. However, these changes and modifications will not depart from the scope of the present specification.

[0033] Figure 2 is a module diagram of an exemplary dose prediction system according to some embodiments of the present specification. As shown in Figure 2 some embodiments, the dose prediction system 200 can include an acquisition module 210, a first prediction module 220, and a second prediction module 230. In some embodiments, the functions corresponding to the dose prediction system 200 can be performed by the processing device 120.

[0034] The acquisition module 210 can be configured to acquire the original dose distribution of the radiotherapy object, the original sign information, and the current sign information. More details about the acquisition of the original dose distribution, the original sign information, and the current sign information can be referred to step 310 and its related description of Figure 3 .

[0035] The first prediction module 220 can be configured to obtain the predicted Pareto optimization parameter by the dose prediction model according to the original dose distribution and the original sign information. More details about the prediction of the Pareto optimization parameter can be referred to step 320 and its related description of Figure 3 .

[0036] The second prediction module 230 can be configured to predict the current dose distribution corresponding to the current sign information using the dose prediction model according to the current sign information and the predicted Pareto optimization parameter. More details about the prediction of the current dose distribution can be referred to step 330 and its related description of Figure 3 .

[0037] It should be understood that Figure 2The system and its modules shown can be implemented in various ways. For example, by hardware, software, or a combination of software and hardware. The system and its modules of the present specification can be implemented not only by hardware circuitry such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but also by software, for example, executed by various types of processors, and by a combination of the above-mentioned hardware circuitry and software (for example, firmware).

[0038] It should be noted that the above description of the system and its modules is for convenience of description and is only illustrative and does not limit the present specification to the embodiments shown. It can be understood that, for those skilled in the art, after understanding the principles of the system, the modules can be combined arbitrarily or connected to form a subsystem without departing from the principles.

[0039] Figure 3 is a flowchart of an exemplary dose prediction method according to some embodiments of the present specification. In some embodiments, the flow 300 can be performed by the processing device 120 or the dose prediction system 200. For example, the flow 300 can be stored in the form of programs or instructions in a storage device (for example, the storage device 140, the storage unit of the processing device 120), and when the processor or Figure 2 The modules shown can implement the flow 300 when executing the programs or instructions. In some embodiments, the flow 300 can utilize one or more additional operations not described below, and / or be completed without one or more operations discussed below. In addition, the order of the operations shown is not limiting. Figure 3 The order of the operations shown is not limiting.

[0040] At step 310, the original dose distribution, the original sign information, and the current sign information of the radiotherapy subject are obtained. In some embodiments, step 310 can be performed by the processing device 120 or the obtaining module 210.

[0041] The dose distribution refers to the radiation dose distributed to each unit volume of the radiotherapy subject. Generally, the radiotherapy process is performed in multiple fractions or stages. Before the entire radiotherapy process, in some embodiments, the doctor can make an initial treatment plan for each treatment fraction in advance, and before the current treatment fraction, the initial treatment plan for the current treatment fraction is adjusted according to the current situation. The original dose distribution can be the dose distribution corresponding to the initial treatment plan of the current treatment fraction. In some embodiments, the original dose distribution can be the dose distribution actually executed by one or more treatment fractions that have been completed. In some embodiments, during the treatment process of the current treatment fraction, changes in the target region and surrounding organs and / or tissues can be monitored in real time by an imaging device (for example, a scanning device in the radiotherapy device 110), and when a large change in the target region or surrounding organs or tissues is observed, the treatment plan currently being executed needs to be adjusted. At this time, the original dose distribution can be the dose distribution corresponding to the treatment plan currently being executed.

[0042] In some embodiments, the original dose distribution can be obtained from the storage device 140, the storage unit of the processing device 120, etc. In some embodiments, the acquisition module can obtain the original dose distribution of the radiotherapy subject by reading from the storage device, database, calling a data interface, etc.

[0043] The physical information is information reflecting the morphology of the organs and / or tissues of the radiotherapy subject. For example, the physical information can reflect the shape, edge, density, etc. of the target region, the critical organ, the normal organ and / or tissue. In some embodiments, the physical information can be represented by a medical image (for example, a CT image, a PET image, an MR image, etc.). For example, the contours of the target region, the critical organ, the normal organ and / or tissue can be outlined in the medical image. In some embodiments, during the radiotherapy process, the physical information of the radiotherapy subject can change. For example, the lesion (target region) becomes smaller during the treatment process, the radiotherapy subject becomes thinner due to poor appetite, etc.

[0044] The original physical information can be the initial physical information of the radiotherapy subject before the entire radiotherapy or the physical information corresponding to the completed treatment fraction.

[0045] In some embodiments, the original physical information can be obtained from the storage device 140, the storage unit of the processing device 120, etc. In some embodiments, the acquisition module can obtain the original physical information of the radiotherapy subject by reading from the storage device, database, calling a data interface, etc.

[0046] The current sign information can be the latest sign information of the radiotherapy subject. For example, the current sign information can be the sign information at the time when the current treatment fraction is about to be performed, the sign information when the current treatment fraction is about to be performed, the sign information in a certain time period (for example, 1 hour ago, 2 hours ago, 5 hours ago, 1 day ago, 2 days ago, 1 week ago, etc.) before the current treatment fraction, or real-time sign information of the radiotherapy subject during the performance of the current treatment fraction.

[0047] In some embodiments, the current sign information can be obtained from the storage device 140, the storage unit of the processing device 120, etc. In some embodiments, a current image of the radiotherapy subject can be obtained from an imaging device (for example, a scanning device in the radiotherapy device 110), and the current sign information of the radiotherapy subject can be obtained by processing the current image. In some embodiments, the current sign information of the radiotherapy subject can be obtained by the obtaining module through reading from a storage device, a database, calling a data interface, etc.

[0048] At step 320, the predicted Pareto optimization parameter is obtained by the dose prediction model according to the original dose distribution and the original sign information. In some embodiments, step 320 can be performed by the processing device 120 or the first prediction module 220.

[0049] At step 330, the current dose distribution corresponding to the current sign information is predicted by the dose prediction model according to the current sign information and the predicted Pareto optimization parameter. In some embodiments, step 330 can be performed by the processing device 120 or the second prediction module 230.

[0050] The Pareto optimization parameter is a parameter corresponding to the optimal solution of a multi-objective optimization problem, for example, a weight parameter of each objective, an optimization target value of each objective, a coordinate of a point on a Pareto plane, or a normal vector of a point on a Pareto plane, etc. In some embodiments, the radiotherapy plan can be regarded as a multi-objective optimization problem, in which each objective conflicts and / or restricts each other. For example, the multiple objectives of the radiotherapy plan can include: the maximum and minimum dose of the target region, the maximum dose per volume, the maximum dose of the protection organ, etc.

[0051] The dose prediction model refers to a machine learning model for realizing radiotherapy dose prediction. The dose prediction model can include, but is not limited to, one or more combinations of a neural network model, a support vector machine model, a k-nearest neighbor model, a decision tree model, etc.

[0052] In some embodiments, the dose prediction model can include a deep neural network model. In some embodiments, the dose prediction model can include at least one of a CNN network, a Res-Net network, and a U-Net network.

[0053] In some embodiments, the input of the dose prediction model includes the original dose distribution, the original sign information, and the current sign information, the dose prediction model obtains a predicted Pareto optimization parameter according to the original dose distribution and the original sign information, and then outputs the current dose distribution according to the current sign information and the predicted Pareto optimization parameter.

[0054] In some embodiments, the dose prediction model includes a first model and a second model. The input of the first model includes the original dose distribution and the original sign information, and the output of the first model includes a predicted Pareto optimization parameter. The input of the second model includes the current sign information and the Pareto optimization parameter output by the first model, and the output of the second model includes the current dose distribution.

[0055] In the process of radiotherapy, although the sign information of the same patient changes, the optimization target is the same, and therefore the Pareto optimization parameter is similar. The predicted Pareto optimization parameter is obtained according to the original sign information and the original dose distribution, and the predicted Pareto optimization parameter is applied to the current fraction of the radiotherapy process, so that a treatment effect similar to the original radiotherapy plan can be achieved.

[0056] The current dose distribution is the radiation dose distributed to each unit volume of the radiotherapy subject corresponding to the current treatment fraction. In some embodiments, the current dose distribution can be represented by a three-dimensional grid. For example, a three-dimensional image of the radiotherapy subject can be divided into 100x100x100 grids, each grid has a volume of 3mmx3mmx3mm, and each grid is labeled as whether it belongs to a target area, a dangerous organ, a normal human tissue, etc., and the corresponding radiation dose of each grid is labeled.

[0057] In some embodiments, the dose prediction model and its training method are described in Figure 4 , Figure 5 and the description thereof.

[0058] It should be noted that the above description of the process 300 is only for example and illustration, and does not limit the scope of the present specification. Various modifications and changes can be made to the process 300 under the guidance of the present specification. However, these modifications and changes are still within the scope of the present specification.

[0059] Figure 4 is a schematic diagram of a first model according to some embodiments of the present specification.

[0060] In some embodiments, the first prediction module 220 can input the original dose distribution and the original sign information into the first model, and output a predicted Pareto optimization parameter corresponding to the original dose distribution and the original sign information from the first model.

[0061] In some embodiments, the first model can include, but is not limited to, a combination of one or more of a neural network model, a support vector machine model, a k-nearest neighbor model, a decision tree model, etc. In some embodiments, the first model can include a deep neural network model. In some embodiments, the first model can include at least one of a CNN network, a Res-Net network, a U-Net network.

[0062] In some embodiments, the first model can be trained based on a plurality of first training samples and labels. For example, each first training sample can include first sign information, a first dose distribution, and a label being a first Pareto optimization parameter corresponding to the first dose distribution.

[0063] In some embodiments, the processing device can obtain a plurality of first training sample sets, each first training sample set including first sign information of a case and at least one set of first Pareto optimization parameters. In some embodiments, the at least one set of first Pareto optimization parameters can be set by artificial or automatic means, or can be set by other means, which are not limited in the present embodiment.

[0064] One form of treatment plan optimization is multi-objective optimization, which is based on an optimization problem including a set of objective functions. According to multi-objective optimization, at least two objective functions used in the optimization problem are incompatible to some extent, because improvement of one objective can result in deterioration of one or more other objectives. For a given solution, if there is no solution that is not worse than it in all objective functions and is better than it in at least one objective function, then the given solution is a Pareto optimal solution. The projection of all Pareto optimal solutions in the objective function space constitutes a Pareto plane. A Pareto optimal solution can correspond to a point on the Pareto plane, and correspond to a Pareto optimization plan and its Pareto optimization parameter. The Pareto optimization parameter can include coordinates of a point on the Pareto plane and / or a normal vector of the point, etc. The Pareto optimization plan can include a dose distribution. By adjusting the Pareto optimization parameter, different points on the Pareto plane can be selected, and thus different Pareto optimization plans can be selected, to find a better dose distribution.

[0065] In some embodiments, the processing device can generate a first Pareto plane according to the first sign information. For example, the processing device can generate the first Pareto plane according to the first sign information by a weighted sum method or an epsilon constraint method. In some embodiments, the processing device can obtain a first dose distribution corresponding to each set of first Pareto optimization parameters on the first Pareto plane. For example, the processing device can find a point corresponding to each set of first Pareto optimization parameters on the first Pareto plane, and determine a corresponding first dose distribution according to a Pareto optimization plan corresponding to the point.

[0066] In some embodiments, the processing device may also obtain a first training sample from historical data. For example, it may obtain vital signs, Pareto optimization parameters, and corresponding dose distributions from historical data for the radiotherapy subject or other patients during their previous (or any) radiotherapy treatment.

[0067] In some embodiments, the processing device may input first vital sign information and first dose distribution into a first initial model to obtain estimated parameters; determine a first loss function based on the estimated parameters and a first Pareto optimization parameter; and update the first initial model based on the first loss function to obtain a dose prediction model.

[0068] For example, such as Figure 4 As shown, the processing device can input the first vital sign information 410 and the first dose distribution 420 into the first initial model to obtain estimated parameters 430; determine the first loss function 440 based on the estimated parameters 430 and the first Pareto optimization parameters; and update the first initial model based on the first loss function 440 to obtain the dose prediction model. In some embodiments, the processing device can perform the above process multiple times to iteratively update the dose prediction model. In some embodiments, in each iteration, the processing device can select one or more first training samples from the first training sample set, and the selection method can be random selection, sequential selection, etc. The first loss function 440 can be the MSE (Mean Square Error) loss function, the cross-entropy loss function, etc., or any combination thereof. The processing device can adjust the parameters of the dose prediction model based on the first loss function to reduce the difference between the estimated parameters 430 and the first Pareto optimization parameters. For example, by continuously adjusting the parameters of the first model, the value of the first loss function can be reduced or minimized.

[0069] Figure 5 This is a schematic diagram of a second model shown according to some embodiments of this specification.

[0070] In some embodiments, the second prediction module 230 can input the current vital signs information and the predicted Pareto optimization parameters into the second model, and the second model can predict the current dose distribution corresponding to the current vital signs information.

[0071] In some embodiments, the second model may be, but is not limited to, a combination of one or more of the following: a neural network model, a support vector machine model, a k-nearest neighbor model, and a decision tree model. In some embodiments, the second model may include a deep neural network model. In some embodiments, the second model may include at least one of a CNN network, a Res-Net network, and a U-Net network.

[0072] In some embodiments, the second model can be trained based on multiple second training samples and labels. For example, each second training sample may include second vital sign information, second Pareto optimization parameters, and a label of a second dose distribution.

[0073] In some embodiments, the processing device may acquire multiple second training sample sets, each second training sample set including second vital sign information of a case and at least one set of second Pareto optimization parameters; in some embodiments, at least one set of second Pareto optimization parameters may be set manually or automatically, or in other ways, and this embodiment does not limit this.

[0074] In some embodiments, the processing device can generate a second Pareto plane based on the second vital sign information. For example, the processing device can generate the second Pareto plane based on the second vital sign information using a weighted sum method or an ε-constraint method. In some embodiments, the processing device can obtain a second dose distribution corresponding to each set of second Pareto optimization parameters on the second Pareto plane. For example, the processing device can find the point corresponding to each set of second Pareto optimization parameters on the second Pareto plane and determine the corresponding second dose distribution based on the Pareto optimization plan corresponding to the point.

[0075] In some embodiments, the processing device may input the second vital sign information and the second Pareto optimization parameters into a second initial model to obtain an estimated dose distribution; determine a second loss function based on the estimated dose distribution and the second dose distribution; and update the second initial model based on the second loss function to obtain a dose prediction model. In some embodiments, the processing device may perform the above process multiple times to iteratively update the dose prediction model. In some embodiments, in each iteration, the processing device may select one or more second training samples from a second training sample set, and the selection method may be random selection, sequential selection, etc.

[0076] For example, such as Figure 5 As shown, the processing device can input the second vital sign information 510 and the second Pareto optimization parameter 520 into the second initial model to obtain the estimated dose distribution 530; determine the second loss function 540 based on the estimated dose distribution 530 and the second dose distribution; update the second initial model based on the second loss function 540 to obtain the dose prediction model. The second loss function 540 can be the MSE (Mean Square Error) loss function, the cross-entropy loss function, or any combination thereof. The processing device can adjust the parameters of the dose prediction model based on the second loss function to reduce the difference between the estimated dose distribution 530 and the second dose distribution. For example, by continuously adjusting the parameters of the second model, the value of the second loss function can be reduced or minimized.

[0077] It is worth noting that in some embodiments, for the at least one first training sample set and the at least one second training sample set, the first sign information and the second sign information can be sign information of different times for the same case, and the at least one first set of Pareto optimization parameters and the at least one second set of Pareto optimization parameters are the same.

[0078] The embodiments of the present specification also provide a dose prediction system, comprising: at least one storage medium storing computer instructions; and at least one processor executing the computer instructions to implement: adjusting a radiotherapy plan based on the current dose distribution and / or generating a new radiotherapy plan.

[0079] In some embodiments of the present specification, (1) the corresponding Pareto optimal parameters are directly located by the original dose distribution, thereby saving the process of adjusting the Pareto optimization parameters by the user, making the automation degree of the adaptive radiotherapy mode higher; (2) the Pareto optimization parameters of the original plan are used for dose prediction under new signs, which can retain the characteristics of the original dose distribution, thereby realizing similar clinical effects as the original dose distribution and improving the accuracy of dose prediction; (3) the time-consuming operation of generating the Pareto plane (for example, the first Pareto plane and the second Pareto plane) is changed from the dose prediction process to the model training process, thereby improving the real-time efficiency of the dose prediction process; (4) by improving the real-time efficiency of the dose prediction process, the optimization speed of the adaptive radiotherapy plan is improved.

[0080] The above has described the basic concept, and it is obvious that the above detailed disclosure is only taken as an example and does not constitute a limitation on the present specification. Although it is not explicitly stated here, those skilled in the art can make various modifications, improvements and corrections to the present specification. Such modifications, improvements and corrections are suggested in the present specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of the present specification.

[0081] Meanwhile, specific words are used in the present specification to describe the embodiments of the present specification. As "one embodiment", "an embodiment", and / or "some embodiments" means a certain feature, structure or characteristic related to at least one embodiment of the present specification. Therefore, it should be emphasized and noted that the "an embodiment" or "one embodiment" or "one alternative embodiment" mentioned in different positions in the present specification does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present specification can be properly combined.

[0082] Furthermore, the order of the processing elements and sequences described in this specification are not intended to be construed as a limitation, unless specifically stated, but are included to provide a complete description of one or more embodiments of the present specification. Regardless of the particular sequence of processing elements and sequences, however, the description herein of a process should be understood to include any and all combinations of one or more elements, and sequences that can be perceived as either open-ended or specific.

[0083] It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting, unless specifically stated as such. It should be noted that, as used in the specification and the appended claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. It should also be noted that, as used in the specification and the appended claims, the term "or" is generally intended to mean "and / or" unless the context clearly dictates otherwise. Furthermore, the words "comprise," "comprising," "include," "including," and the like are generally intended to be synonymous, unless the context clearly dictates otherwise. Unless specifically stated otherwise, and as can be apparent from the disclosure, use of terms such as "processing," "computing," "calculating," "determining," "displaying," or the like, refer to actions or processes of a machine that manipulates or transforms data represented as physical electronic or magnetic quantities within memories, registers, or other information storage devices. Note, too, that while embodiments can be implemented by software, the software can be written in any of numerous languages or combinations of languages, and can be executed in a machine or on a machine.

[0084] Some embodiments use numerical values to describe components, quantities of attributes, and the like. It should be understood that such numerical values used in the description of embodiments are, in some examples, modified by the adjectives "about," "approximately," or "generally." Unless otherwise stated, "about," "approximately," or "generally" indicates that a value can vary by ±20%. Accordingly, numerical values used in the specification and claims are approximations that vary depending on the requirements of the particular embodiment. In some embodiments, numerical values should be considered to be defined with the specified degree of accuracy and with the normal rounding of numbers to the nearest significant figure. Although the numerical ranges and parameters setting forth the broadest scope of the embodiments described herein are approximations, the numerical values set forth in the specific examples are reported as precisely as practicable. The numerical values set forth in the specific examples are provided to be as precise as reasonably possible. However, some variations may

[0085] Each patent, patent application, patent publication, and other material, such as articles, books, specifications, publications, documents, and the like, referenced herein are hereby incorporated by reference in their entirety for the teachings relevant to the sentence and / or paragraph in which such reference is made. Discrepancies between applications history documents and the present specification, other than limitations on the scope of the present claims, are excepted, as are documents that can be added to the present specification after filing (present or future additions to the specification). Note that if there is a discrepancy between descriptions, definitions, and / or terminology in the present specification and in the incorporated material, the description, definitions, and / or terminology in the present specification control.

[0086] Finally, it should be understood that the embodiments described herein are only given by way of example and that other modifications can occur to persons skilled in the art. Therefore, the scope of the present description is not intended to be limited to the embodiments described herein but is only limited by the claims that follow.

Claims

1. A dose prediction method characterized by, The method comprises: obtaining an original dose distribution, original sign information and current sign information of a radiotherapy object; inputting the original dose distribution and the original sign information into a first model to output a predicted Pareto optimization parameter, the first model being a machine learning model; inputting the current sign information and the predicted Pareto optimization parameter into a second model to output a current dose distribution corresponding to the current sign information.

2. The method of claim 1, wherein, The training process of the first model comprises: obtaining a plurality of first training sample sets, each first training sample set comprising first sign information of a case and at least one set of first Pareto optimization parameters; generating a first Pareto plane according to the first sign information; obtaining a first dose distribution corresponding to each set of first Pareto optimization parameters on the first Pareto plane; inputting the first sign information and the first dose distribution into a first initial model to obtain an estimated parameter; determining a first loss function according to the estimated parameter and the first Pareto optimization parameter; updating the first initial model according to the first loss function to obtain the first model.

3. The method of claim 2, wherein, The training process of the second model comprises: obtaining a plurality of second training sample sets, each second training sample set comprising second sign information of a case and at least one set of second Pareto optimization parameters; generating a second Pareto plane according to the second sign information; obtaining a second dose distribution corresponding to each set of second Pareto optimization parameters on the second Pareto plane; inputting the second sign information and the second Pareto optimization parameter into a second initial model to obtain an estimated dose distribution; determining a second loss function according to the estimated dose distribution and the second dose distribution; updating the second initial model according to the second loss function to obtain the second model.

4. The method of claim 3, wherein, For at least one of the first training sample sets and at least one of the second training sample sets, the first sign information and the second sign information are sign information of the same case at different times, and the at least one set of first Pareto optimization parameters and the at least one set of second Pareto optimization parameters are the same.

5. The method of claim 1, wherein, The first model and / or the second model comprise one or more combinations of a neural network model, a support vector machine model, a k-nearest neighbor model, a decision tree model, etc.

6. A dose prediction system characterized in that, The system comprises: an acquisition module configured to obtain an original dose distribution, original sign information and current sign information of a radiotherapy object; a first prediction module configured to input the original dose distribution and the original sign information into a first model to output a predicted Pareto optimization parameter, the first model being a machine learning model; a second prediction module configured to input the current sign information and the predicted Pareto optimization parameter into a second model to output a current dose distribution corresponding to the current sign information.

7. The system of claim 6, wherein, The system comprises: at least one storage medium storing computer instructions; and at least one processor executing the computer instructions to implement: adjusting a radiotherapy plan based on the current dose distribution and / or generating a new radiotherapy plan.

8. A dose prediction apparatus characterized by comprising: The device comprises: at least one storage medium storing computer instructions; at least one processor to execute the computer instructions to implement the method of any one of claims 1-5.

9. A computer-readable storage medium storing computer instructions, which, when read by a computer, cause the computer to perform the method of any one of claims 1-5.

Images