Dose distribution prediction method and system, plan generation method and system and storage medium

Abstract

The invention provides a dose distribution prediction method and system, a plan generation method and system and a storage medium, relates to the technical field of radiotherapy, and is used for solving the problem that the accuracy of predicted dose distribution is relatively low. The method comprises the steps that a prediction device obtains a target volume image, a structural mask image of a target volume and multiple radiation field direction features in different radiation field directions, wherein the target volume comprises a target area and organs at risk; and then, the prediction device inputs the target volume image, the structural mask image of the target volume and the multiple radiation field direction features of different radiation field directions into the trained dose prediction model to obtain the dose distribution of the target volume.

Description

Technical Field

[0001] The present application relates to the field of radiotherapy technology, and in particular to a dose distribution prediction method, a plan generation method, a system and a storage medium. Background Art

[0002] With the continuous development of technology, radiation can be used to irradiate the tumor area, thereby eliminating the primary or metastatic lesions of local tumors. Currently, making radiotherapy plans is relatively time-consuming, and professional radiation therapy physicists are required to continuously optimize and adjust the radiation dose distribution to ensure that the radiation dose to surrounding normal tissues is reduced as much as possible while ensuring the local control rate of the tumor.

[0003] Generally, each tumor patient has unique disease manifestations, and planners need to develop a corresponding treatment plan for each tumor patient. In order to improve the efficiency of treatment planning, the radiotherapy planning system can predict the dose distribution based on the object's structural information. However, the current prediction scheme can only be performed based on the object's structural information, and the amount of reference information is relatively small. This may result in a low accuracy of the predicted dose distribution. Summary of the invention

[0004] The present application provides a dose distribution prediction method, a plan generation method, a system and a storage medium, which are used to solve the problem of low accuracy of predicted dose distribution.

[0005] In order to achieve the above objectives, this application adopts the following technical solutions:

[0006] In a first aspect, the present application provides a method for predicting dose distribution. In the method, a device for predicting dose distribution (which may be referred to as a "prediction device") obtains a target volume image, a structural mask image of the target volume, and field direction features of multiple different field directions, where the target volume includes a target area and organs at risk. Afterwards, the prediction device inputs the target volume image, the structural mask image of the target volume, and the field direction features of multiple different field directions into a trained dose prediction model to obtain a dose distribution of the target volume.

[0007] Based on the above technical solution, the prediction device can obtain the target volume image, the structural mask image of the target volume and the field direction characteristics of multiple different field directions, and the target volume includes the target area and the organs at risk. The prediction device can input the target volume image, the structural mask image of the target volume and the field direction characteristics of multiple different field directions into the trained dose prediction model to obtain the dose distribution of the target volume. In other words, the predicted dose distribution is based on the target area, the organs at risk and the field direction. In this way, when performing dose prediction, the real planned field information is introduced, which reduces the uncertainty of the dose distribution prediction, thereby improving the accuracy of the predicted dose distribution.

[0008] In a possible design, the field direction characteristics of multiple different field directions include: multiple light path characteristic images of different field directions.

[0009] In another possible design, the field direction characteristics of the multiple field directions include: field direction codes of multiple different field directions.

[0010] In another possible design, the characteristic images of the light paths of the multiple different field directions are obtained by calculating the actual light path passing through the target volume in each of the multiple different field directions based on a ray tracing algorithm.

[0011] In a second aspect, the present application provides a plan generation method, the method comprising:

[0012] The plan generating device (which may be referred to as "generating device") may obtain a target volume image, a structural mask image of the target volume, and field direction features of multiple different field directions, wherein the target volume includes a target area and organs at risk. Afterwards, the generating device inputs the target volume image, the structural mask image of the target volume, and field direction features of multiple different field directions into the trained dose prediction model to obtain the dose distribution of the target volume. Next, the generating device may construct a dose volume histogram based on the dose distribution of the target volume. Then, the generating device may use the dose volume histogram as an objective function, perform inverse optimization, and generate a treatment plan.

[0013] Based on the above technical solution, the generating device can obtain the target volume image, the structural mask image of the target volume and the field direction features of multiple different field directions, and the target volume includes the target area and the organs at risk. Afterwards, the generating device can input the target volume image, the structural mask image of the target volume and the field direction features of multiple different field directions into the trained dose prediction model to obtain the dose distribution of the target volume, and construct a dose volume histogram based on the dose distribution of the target volume. Then, the generating device can use the dose volume histogram as the objective function, perform inverse optimization, and generate a treatment plan. In this way, the generating device can predict an accurate dose distribution based on the target area, organs at risk, and field direction, and then guide the optimization of the treatment plan with an accurate dose distribution, which can improve the accuracy of the treatment plan and the efficiency of formulating the treatment plan.

[0014] In a possible design, the field direction characteristics of multiple different field directions include: multiple light path characteristic images of different field directions.

[0015] In another possible design, the field direction characteristics of the multiple field directions include: field direction codes of multiple different field directions.

[0016] In another possible design, the characteristic images of the light paths of the multiple different field directions are obtained by calculating the actual light path passing through the target volume in each of the multiple different field directions based on a ray tracing algorithm.

[0017] In a third aspect, the present application provides a computer system comprising: a processor and a memory; the processor and the memory are coupled; the memory is used to store one or more programs, the one or more programs comprising computer execution instructions, and when the computer system is running, the processor executes the computer execution instructions stored in the memory to implement the dose distribution prediction method as described in the first aspect, the second aspect and any possible implementation thereof.

[0018] In a fourth aspect, the present application provides a computer-readable storage medium, which stores instructions. When the instructions are executed on a computer, the computer executes the method described in the first aspect, the second aspect and any possible implementation thereof.

[0019] In the above scheme, the technical problems that can be solved by the computer system and computer storage medium and the technical effects achieved can be referred to the technical problems and technical effects solved by the first and second aspects above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A schematic diagram of a flow chart of a method for predicting dose distribution provided in an embodiment of the present application;

[0021] Figure 2 An example diagram of a light path image provided in an embodiment of the present application;

[0022] Figure 3 A schematic flow chart of another method for predicting dose distribution provided in an embodiment of the present application;

[0023] Figure 4 A schematic diagram of a plan generation method provided in an embodiment of the present application;

[0024] Figure 5 A schematic diagram of the structure of a device provided in an embodiment of the present application;

[0025] Figure 6 A conceptual partial view of a computer program product provided for an embodiment of the present application. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0027] In this article, the character " / " generally indicates that the objects before and after are in an "or" relationship. For example, A / B can be understood as A or B.

[0028] The terms “first” and “second” in the description and claims of the present application are used to distinguish different objects rather than to describe a specific order of the objects.

[0029] In addition, the terms "including" and "having" and any variations thereof mentioned in the description of this application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or modules is not limited to the listed steps or modules, but may optionally include other steps or modules that are not listed, or may optionally include other steps or modules that are inherent to these processes, methods, products or devices.

[0030] In addition, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present concepts in a specific way.

[0031] Before introducing the dose distribution prediction method of the embodiment of the present application in detail, the implementation environment and application scenarios of the embodiment of the present application are first introduced.

[0032] First, the application scenarios of the embodiments of the present application are introduced.

[0033] The dose distribution prediction method of the embodiment of the present application is applied to the scenario of predicting dose distribution. In recent years, artificial intelligence has shown great potential in automatic radiotherapy planning tasks. For example, organ at risk segmentation, target area definition, and dose distribution estimation. Among them, organ at risk segmentation is relatively mature and has entered the clinic to assist radiotherapists in rapid contouring, greatly reducing the workload of staff. At the same time, three-dimensional dose distribution prediction is also gradually being carried out in the clinic as a standardized method, assisting planners to save time in making plans, and improving work efficiency and plan quality. The three-dimensional dose prediction model is based on a deep learning method, which takes the patient's anatomical information (target area and organ at risk contours, CT images and other features) as the model input, and the predicted three-dimensional dose distribution as the model output.

[0034] However, in the above technical solution, the predicted three-dimensional dose distribution only uses the patient's anatomical information as input, and the predicted three-dimensional dose distribution deviates greatly from the actual three-dimensional dose distribution, resulting in low accuracy of the predicted three-dimensional dose distribution. Therefore, how to improve the accuracy and certainty of the predicted dose distribution has become a technical problem that needs to be solved urgently.

[0035] In order to solve the above problems, an embodiment of the present application provides a method for predicting dose distribution, wherein a prediction device can obtain a target volume image, a structural mask image of the target volume, and field direction features of multiple different field directions, wherein the target volume includes a target area and organs at risk. The prediction device can input the target volume image, the structural mask image of the target volume, and field direction features of multiple different field directions into a trained dose prediction model to obtain the dose distribution of the target volume. In other words, the predicted dose distribution is obtained based on the target area, organs at risk, and field direction. In this way, when performing dose prediction, real planned field information is introduced, which reduces the uncertainty of dose distribution prediction, thereby improving the accuracy and certainty of the predicted dose distribution. Moreover, by predicting dose distribution through a dose prediction model, the efficiency of predicting dose distribution can be improved.

[0036] The embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0037] like Figure 1 As shown, a method for predicting dose distribution provided in an embodiment of the present application includes:

[0038] S101. A prediction device obtains a target volume image, a structural mask image of the target volume, and field direction features of multiple different field directions.

[0039] The target volume includes the target area and the organ at risk. The target volume image includes the image of the target area and the image of the organ at risk.

[0040] Exemplarily, the target volume image may be a computed tomography (CT) image, a magnetic resonance imaging (MRI) image, a positron emission tomography (PET) image, or the like.

[0041] In a possible implementation manner, after the prediction device acquires the target volume image, the prediction device may process the target volume image to obtain a structural mask image of the target volume.

[0042] It should be noted that the structural mask image of the target volume can also be obtained before the prediction device acquires the target volume image, for example, the target volume image can be processed by other image processing devices / software to obtain the structural mask image of the target volume. For example, the structural mask image of the target volume can be obtained by using a treatment planning system or a contouring system.

[0043] Exemplarily, the structural mask image of the target volume includes: a target volume area, a non-target volume area, and a contour line. The target volume area is within the contour line, and the non-target volume is outside the contour line. The image pixel value within the contour line is 1, and the image pixel value outside the contour line is 0.

[0044] It should be noted that, in the embodiment of the present application, the field direction feature may be field direction data, or field direction code, or processed field direction data. The field direction data is a specific field direction, such as 0 degrees, 40 degrees, 80 degrees, and the like.

[0045] In a possible implementation, the field direction feature may be field direction data. The prediction device may obtain field direction data of multiple different field directions.

[0046] In a possible design, the field direction data is set by a user. The prediction device may receive a first operation of the user, the first operation being used to input the field direction data. In response to the first operation, the prediction device acquires field direction data of a plurality of different field directions.

[0047] In another possible implementation, the field direction features of the plurality of different field directions include: field direction codes of the plurality of different field directions, the field direction codes being used to indicate field direction data. The prediction device may receive a second operation of the user, the second operation being used to input the field direction code. In response to the second operation, the prediction device acquires the field direction codes of the plurality of different field directions.

[0048] In another possible implementation, the field direction feature is processed field direction data. After acquiring the field direction data of multiple different field directions, the prediction device processes the field direction data of multiple different field directions to obtain the field direction features of multiple different field directions.

[0049] In a possible design, the projection direction characteristics of a plurality of different projection directions include: a plurality of light path characteristic images of a plurality of different projection directions.

[0050] In the embodiment of the present application, the characteristic images of the light paths of the multiple different field directions are obtained by calculating the actual light paths passing through the target volume in each of the multiple different field directions based on a ray tracing algorithm. Specifically, the prediction device can perform ray tracing processing on the field direction data of each of the multiple different field directions to obtain characteristic images of the light paths of the multiple different field directions.

[0051] For example, Figure 2 As shown, the irradiation path of light A is path 201. The irradiation path of light B is path 202. Among them, the pixels of the area where the light passes through are represented by 1, and the pixels that do not pass through the area are represented by 0.

[0052] Optionally, the characteristic images of light paths in a plurality of different projection directions further include: an irradiation area of ​​the light in each projection direction.

[0053] S102, the prediction device inputs the target volume image, the structural mask image of the target volume and the field direction features of multiple different field directions into the trained dose prediction model to obtain the dose distribution of the target volume.

[0054] Optionally, the dose distribution may be a three-dimensional dose distribution.

[0055] It should be noted that three-dimensional dose distribution refers to the distribution of radiotherapy dose in three-dimensional space. It can be used to evaluate the effect of radiotherapy and optimize radiotherapy plans. There are many forms of three-dimensional dose distribution, such as isodose lines, dose volume maps, dose surface maps, etc. Usually, three-dimensional dose distribution can be represented by isodose lines, which can intuitively show the distribution of radiotherapy dose in different directions.

[0056] In a possible implementation, the prediction device may input the target volume image, the structural mask image of the target volume, and the light path characteristic images of multiple different radiation field directions into the trained dose prediction model to obtain the dose distribution of the target volume.

[0057] For example, Figure 3As shown, the prediction device can input the target volume image 301, the structural mask image 302 of the target volume, and the light path characteristic images 303 of multiple different field directions into the input layer of the trained dose prediction model. Afterwards, the input layer inputs the target volume image 301, the structural mask image 302 of the target volume, and the light path characteristic images 303 of multiple different field directions into the hidden layer (i.e., the dose prediction network) to obtain the dose distribution of the target volume. Afterwards, the output layer outputs the dose distribution of the target volume.

[0058] In another possible implementation, the prediction device may input the target volume image, the structural mask image of the target volume, and the field direction encodings of multiple different field directions into the trained dose prediction model to obtain the dose distribution of the target volume.

[0059] In another possible implementation, the prediction device may input the target volume image, the structural mask image of the target volume, and the field direction data of multiple different field directions into the trained dose prediction model to obtain the dose distribution of the target volume.

[0060] It should be noted that the embodiments of the present application do not limit the predicted dose model. For example, the predicted dose model is built based on a U-net network or a DenseNet network. For a specific description of the process of training the predicted dose model by the prediction device to obtain the trained predicted dose model, reference can be made to the introduction of the training model in the conventional technology, which will not be repeated here.

[0061] Based on the above technical solution, the prediction device can obtain the target volume image, the structural mask image of the target volume and the field direction characteristics of multiple different field directions, and the target volume includes the target area and the organs at risk. The prediction device can input the target volume image, the structural mask image of the target volume and the field direction characteristics of multiple different field directions into the trained dose prediction model to obtain the dose distribution of the target volume. In other words, the predicted dose distribution is obtained based on the target area, the organs at risk and the field direction. In this way, when performing dose prediction, the real planned field information is introduced, which reduces the uncertainty of the dose distribution prediction, thereby improving the accuracy and certainty of the predicted dose distribution.

[0062] In some embodiments, a plurality of trained preset prediction models are deployed in the prediction device, and the plurality of preset prediction models include a dose prediction model. The prediction device may receive a third operation of the user, and the third operation is used to input parameter information, and the parameter information includes at least one of the following: treatment equipment, irradiation technology, total dose, number of treatments, etc. Thereafter, in response to the third operation, the prediction device may determine a dose prediction model from at least one preset prediction model based on the parameter information. One preset prediction model corresponds to a set of parameter information.

[0063] For example, if the treatment device is device a, the irradiation technology is intensity modulation technology, the total dose is 30, and the number of treatments is 5, then the corresponding preset prediction model is model a. If the treatment device is device b, the irradiation technology is intensity modulation technology, the total dose is 50, and the number of treatments is 10, then the corresponding preset prediction model is model b.

[0064] It is understandable that, through parameter information (such as treatment equipment, irradiation technology, total dose, number of treatments), the prediction device can determine a matching dose prediction model, thereby improving the accuracy of the dose prediction model in predicting dose distribution.

[0065] The following is an introduction to an embodiment of the present application by taking the example that the field direction characteristics of multiple different field directions include multiple light path characteristic images of different field directions, that is, the parameters of the dose prediction model after input training include: a target volume image, a structural mask image of the target volume and multiple light path characteristic images of different field directions.

[0066] In some embodiments, before acquiring a target volume image, a structural mask image of the target volume, and a plurality of light path characteristic images of different projection field directions, the prediction device may acquire a first image and a second image, wherein the first image includes: an image of the target volume, and the second image is a structural mask image of the first image. Afterwards, the prediction device may interpolate the first image and the second image to the same resolution as the preset dose grid according to the size of the preset dose grid. Afterwards, the prediction device may use the first image with the same resolution as the preset dose grid as the target volume image, and use the second image with the same resolution as the preset dose grid as the structural mask image of the target volume.

[0067] That is, the target volume image and the structure mask image of the target volume have the same resolution as the preset dose grid.

[0068] It should be noted that the first image is the target area and the outline of the organ at risk drawn by the prediction device in the treatment planning system according to the CT image. After that, after confirmation by the user (such as inputting a confirmation operation), the definition of the target area and the structure and outline of the organ at risk is determined to obtain the first image.

[0069] In some embodiments, the prediction device may also obtain a plurality of first light path images in different projection directions. Afterwards, the prediction device may normalize the pixels of the first image, the pixels of the second image, and the pixels of the first light path images in different projection directions to obtain a target volume image, a structural mask image of the target volume, and a plurality of light path feature images in different projection directions. Among them, the pixels of the target volume image are smaller than the pixels of the first image, the pixels of the structural mask image of the target volume are smaller than the pixels of the second image, and the pixels of the light path feature image are smaller than the pixels of the first light path image.

[0070] That is to say, in the embodiment of the present application, the target volume image, the structural mask image of the target volume, and the light path characteristic images of multiple different projection directions are images after pixel normalization.

[0071] It is understandable that the prediction device can reduce the pixels of the target volume image, the structure mask image of the target volume, and the light path feature images of multiple different field directions by normalizing the pixels of multiple images, thereby facilitating the convergence of the model.

[0072] Optionally, after the prediction device can obtain the first image and the second image, the prediction device can obtain the center point of the target volume, and adjust both the first image and the second image to a preset image size according to the center point.

[0073] That is to say, in the embodiment of the present application, the image sizes of the target volume image and the structural mask image of the target volume are the same.

[0074] It should be noted that, in some embodiments, the prediction device may obtain the target volume image, the structure mask image of the target volume and the field direction features of multiple different field directions from the first device. The first device is a device other than the prediction device.

[0075] In a possible implementation, the first device may acquire a first image, a second image, and a plurality of first light path images of different field directions. Afterwards, the first device may process the first image, the second image, and a plurality of first light path images of different field directions, respectively, to obtain a target volume image, a structural mask image of the target volume, and field direction features of a plurality of different field directions. Afterwards, the first device may send the target volume image, the structural mask image of the target volume, and the field direction features of a plurality of different field directions to the prediction device, and the prediction device may receive the target volume image, the structural mask image of the target volume, and the field direction features of a plurality of different field directions from the first device.

[0076] It should be noted that the process of the first device processing the first image, the second image and the first light path images of multiple different projection directions respectively can refer to the description of the preset device processing the first image, the second image and the first light path images of multiple different projection directions in the above embodiment, and will not be repeated here.

[0077] In some embodiments, after acquiring the target volume image, the structural mask image of the target volume, and the characteristic images of the light paths in multiple different projection directions, the prediction device may input the target volume image, the structural mask image of the target volume, and the characteristic images of the light paths in multiple different projection directions into the trained dose prediction model, and the dose prediction model performs channel superposition processing on the target volume image, the structural mask image of the target volume, and the characteristic images of the light paths in multiple different projection directions to determine the target tensor. Afterwards, the prediction device analyzes the target tensor through the dose prediction model to obtain the dose distribution of the target volume.

[0078] Optionally, the prediction device can perform channel superposition processing on the target volume image, the structural mask image of the target volume, and the light path characteristic images of multiple different radiation field directions to determine the target tensor. Afterwards, the prediction device inputs the target tensor into the trained dose prediction model to obtain the dose distribution of the target volume.

[0079] The following is an introduction to a method for generating a treatment plan based on the dose distribution in the above embodiment.

[0080] like Figure 4 , is a plan generation method provided in an embodiment of the present application, the method comprising:

[0081] S401. A generating device obtains a target volume image, a structural mask image of the target volume, and field direction features of a plurality of different field directions.

[0082] The target volume includes the target area and organs at risk.

[0083] In a possible design, the projection direction characteristics of a plurality of different projection directions include: a plurality of light path characteristic images of a plurality of different projection directions.

[0084] In an embodiment of the present application, the characteristic images of light paths in multiple different field directions are obtained by calculating the actual light path passing through the target volume in each of the multiple different field directions based on a ray tracing algorithm.

[0085] In another possible design, the field direction characteristics of the multiple field directions include: field direction codes of multiple different field directions.

[0086] It should be noted that, for a specific introduction to the generation device acquiring the target volume image, the structural mask image of the target volume and the field direction characteristics of multiple different field directions, reference may be made to the description of S101 in the above embodiment, which will not be repeated here.

[0087] S402, the generating device inputs the target volume image, the structural mask image of the target volume and the field direction features of multiple different field directions into the trained dose prediction model to obtain the dose distribution of the target volume.

[0088] It should be noted that, for the introduction of the generating device inputting the target volume image, the structural mask image of the target volume and the field direction features of multiple different field directions into the trained dose prediction model to obtain the dose distribution of the target volume, reference can be made to the description of S102 in the above embodiment, which will not be repeated here.

[0089] S403: The generating device constructs a dose-volume histogram according to the dose distribution of the target volume.

[0090] In a possible implementation, the generating device uses a segmented uniform sampling method to construct a dose volume histogram based on the dose distribution of the target volume.

[0091] In another possible implementation manner, the generating device performs hierarchical clustering based on the dose distribution of the target volume to construct a dose volume histogram.

[0092] In another possible implementation, the generating device may process the dose distribution of the target volume based on a deep learning algorithm to construct a dose volume histogram.

[0093] S404: The generating device uses the dose-volume histogram as an objective function, performs inverse optimization, and generates a treatment plan.

[0094] In a possible implementation, the generating device may use the dose volume histogram as an objective function, optimize the flux map by the objective function, and obtain an optimized flux map. Afterwards, the generating device may optimize the preset plan according to the optimized flux map to obtain an optimized treatment plan.

[0095] In one possible design, the treatment plan can be used to guide radiation delivery by a radiotherapy device.

[0096] In some embodiments, the radiotherapy device includes: a radiation source for emitting a therapeutic beam, a frame for carrying the radiation source, and a beam collimation unit for regulating the shape of the beam. The treatment plan is used to guide the irradiation position, irradiation duration, movement trajectory of the radiation source, size / shape / position of the beam collimation unit, etc. Under the guidance of the treatment plan, the radiotherapy device delivers the radiation dose specified in the treatment plan to the target area.

[0097] Based on the above technical solution, the generating device can obtain the target volume image, the structural mask image of the target volume and the field direction features of multiple different field directions, and the target volume includes the target area and the organs at risk. Afterwards, the generating device can input the target volume image, the structural mask image of the target volume and the field direction features of multiple different field directions into the trained dose prediction model to obtain the dose distribution of the target volume, and construct a dose volume histogram based on the dose distribution of the target volume. Then, the generating device can use the dose volume histogram as the objective function to perform inverse optimization and generate a treatment plan. In this way, the generating device can optimize the dose distribution obtained through the target area, the organs at risk and the field direction to obtain a treatment plan. In this way, when performing radiotherapy, a treatment plan can be formulated based on the introduction of real planned field information, thereby improving the accuracy of the treatment plan and the efficiency of determining the plan.

[0098] It should be noted that the execution subject of S101-S102 and S401-S404 can be the same, that is, the prediction device and the generation device are the same device. For example, the prediction device can execute S101-S102 and S401-S404. For another example, the generation device can execute S101-S102 and S401-S404. Alternatively, the execution subject of S101-S102 and S401-S404 is different, that is, the prediction device and the generation device are different devices. This embodiment of the application is not limited to this.

[0099] The above mainly introduces the solution provided by the embodiment of the present application from the perspective of the method. It is understandable that the dose distribution prediction device, or the plan generation device or the server includes hardware structures and / or software modules corresponding to the execution of each function in order to realize the above functions. Those skilled in the art should easily realize that, in combination with the steps of the dose distribution prediction method described in each example of the embodiment disclosed in the present application, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present application.

[0100] The embodiment of the present application also provides a device for predicting dose distribution. The device for predicting dose distribution can be a computer device, a CPU in the above-mentioned computer device, a prediction module in the above-mentioned computer device for predicting dose distribution, or a client in the above-mentioned computer device for predicting dose distribution.

[0101] Similarly, the embodiment of the present application also provides a plan generation device. The plan generation device can be a computer device, a CPU in the above computer device, a generation module in the above computer device for generating a plan, or a client in the above computer device for generating a plan.

[0102] The embodiment of the present application can divide the dose distribution prediction device (or plan generation device) into functional modules or functional units according to the above method example. For example, each functional module or functional unit can be divided according to each function, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware or in the form of software functional modules or functional units. Among them, the division of modules or units in the embodiment of the present application is schematic, which is only a logical functional division, and there may be other division methods in actual implementation.

[0103] Figure 5 1 is a schematic diagram of a hardware structure of a device according to an exemplary embodiment, and the device may be a device for predicting dose distribution or a device for generating a plan. The device may include a processor 502, and the processor 502 is used to execute application code to implement the method for predicting dose distribution and / or the method for generating a plan in the present application.

[0104] The processor 502 may be a central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the present application.

[0105] like Figure 5 As shown, the device may further include a memory 503. The memory 503 is used to store application code for executing the solution of the present application, and is controlled to execute by the processor 502.

[0106] The memory 503 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 503 may exist independently and be connected to the processor 502 via the bus 504. The memory 503 may also be integrated with the processor 502.

[0107] like Figure 5 As shown, the device for predicting dose distribution may further include a communication interface 501, wherein the communication interface 501, the processor 502, and the memory 503 may be coupled to each other, for example, via a bus 504. The communication interface 501 is used to exchange information with other devices, for example, to support information exchange between the device for predicting dose distribution and other devices.

[0108] It should be pointed out that Figure 5 The device structure shown in the figure does not constitute a limitation on the dose distribution prediction device or plan generation device, except Figure 5 In addition to the components shown, the dose distribution prediction device and the plan generation device may include more or fewer components than shown, or combine certain components, or have a different arrangement of components.

[0109] The present application also provides a computer system, which includes: a processor and a memory; the processor and the memory are coupled; the memory is used to store one or more programs, and the one or more programs include computer execution instructions. When the computer system is running, the processor executes the computer execution instructions stored in the memory to implement the dose distribution prediction method and / or plan generation method in the present application.

[0110] The present application also provides a computer-readable storage medium, on which instructions are stored. When the instructions in the computer-readable storage medium are executed by a processor of a computer device, the computer can execute the dose distribution prediction method and plan generation method provided in the above-mentioned embodiments. For example, the computer-readable storage medium may be a memory 503 including instructions, and the above instructions may be executed by a processor 502 of a computer device to complete the above method. Optionally, the computer-readable storage medium may be a non-temporary computer-readable storage medium, for example, a non-temporary computer-readable storage medium may be a ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.

[0111] Figure 6 A conceptual partial view of a computer program product provided by an embodiment of the present application is schematically shown, where the computer program product includes a computer program for executing a computer process on a computing device.

[0112] In one embodiment, the computer program product is provided using a signal bearing medium 600. The signal bearing medium 600 may include one or more program instructions that, when executed by one or more processors, may provide the above-described Figure 1 and Figure 4 Thus, for example, reference to Figure 1 In the embodiment shown in , one or more features of S101-S102 may be undertaken by one or more instructions associated with the signal bearing medium 600. In addition, Figure 6 The program instructions in also describe example instructions.

[0113] In some examples, the signal bearing medium 600 may include a computer readable medium 601, such as, but not limited to, a hard drive, a compact disk (CD), a digital video disk (DVD), a digital tape, a memory, a read-only memory (ROM), or a random access memory (RAM), and the like.

[0114] In some implementations, signal bearing medium 600 may include computer recordable medium 602 such as, but not limited to, memory, read / write (R / W) CD, R / W DVD, and the like.

[0115] In some implementations, signal bearing medium 600 may include communication medium 603 such as, but not limited to, digital and / or analog communication media (eg, fiber optic cables, waveguides, wired communication links, wireless communication links, etc.).

[0116] The signal bearing medium 600 may be communicated by a wireless form of communication medium 603. The one or more program instructions may be, for example, computer executable instructions or logic implemented instructions.

[0117] In some examples, such as for Figure 5 The described apparatus (eg, the dose distribution prediction apparatus) may be configured to provide various operations, functions, or actions in response to one or more program instructions via computer-readable medium 601 , computer-recordable medium 602 , and / or communication medium 603 .

[0118] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0119] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0120] The units described as separate components may or may not be physically separated, and the components shown as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0121] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0122] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or the full classification part or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium, including a number of instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor (processor) to execute the full classification part or part of the steps of each embodiment method of the present application. The aforementioned storage medium includes various media that can store program codes, such as-U disk, mobile hard disk, ROM, RAM, disk or CD.

[0123] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto, and any changes or substitutions within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for predicting dose distribution, characterized in that: The method comprises: Acquiring a target volume image, a structural mask image of the target volume, and field direction features of a plurality of different field directions, wherein the target volume includes a target area and an organ at risk; The target volume image, the structural mask image of the target volume and the field direction features of the multiple different field directions are input into the trained dose prediction model to obtain the dose distribution of the target volume.

2. The method for predicting dose distribution according to claim 1, characterized in that: The projection direction characteristics of the plurality of different projection directions include: light path characteristic images of the plurality of different projection directions.

3. The method for predicting dose distribution according to claim 1, characterized in that: The field direction characteristics of the multiple field directions include: field direction codes of the multiple different field directions.

4. The method for predicting dose distribution according to claim 2, characterized in that: The light path characteristic images of the multiple different field directions are obtained by calculating the actual light path passing through the target volume in each of the multiple different field directions based on a ray tracing algorithm.

5. A plan generation method, characterized in that: The method comprises: Acquiring a target volume image, a structural mask image of the target volume, and field direction features of a plurality of different field directions, wherein the target volume includes a target area and an organ at risk; Inputting the target volume image, the structural mask image of the target volume and the field direction features of the multiple different field directions into the trained dose prediction model to obtain the dose distribution of the target volume; constructing a dose volume histogram according to the dose distribution of the target volume; The dose volume histogram is used as the objective function to perform inverse optimization and generate a treatment plan.

6. The plan generation method according to claim 5, characterized in that: The projection direction characteristics of the plurality of different projection directions include: light path characteristic images of the plurality of different projection directions.

7. The plan generation method according to claim 5, characterized in that: The field direction characteristics of the multiple field directions include: field direction codes of the multiple different field directions.

8. The plan generation method according to claim 6, characterized in that: The light path characteristic images of the multiple different field directions are obtained by calculating the actual light path passing through the target volume in each of the multiple different field directions based on a ray tracing algorithm.

9. A computer system, characterized in that: include: Processor and memory; The processor is coupled to the memory; The memory is used to store one or more programs, which include computer-executable instructions. The processor executes the computer-executable instructions stored in the memory to enable the computer system to perform the method according to any one of claims 1 to 4.

10. A computer-readable storage medium, wherein instructions are stored in the computer-readable storage medium, characterized in that: When a computer executes the instruction, the computer performs the method according to any one of claims 1 to 4.

Images