Radiotherapy plan generation method, system, device and medium based on ordinary differential equations

Abstract

The present invention discloses a method, system, device, and medium for generating radiotherapy plans based on ordinary differential equations. This method relates to the automatic generation of radiotherapy plans in the field of artificial intelligence medicine, and aims to address the technical issues of low dose distribution accuracy and poor radiotherapy quality in radiotherapy plans generated in the prior art. This method employs a dose prediction module and a flux prediction module within the radiotherapy plan generation model, both of which utilize a U-Net network structure. The projection map is used as input to the dose prediction module, and both the output of the dose prediction module and the projection map are used as input to the flux prediction module. When the flux prediction module is upsampled, the feature map obtained after upsampling, the feature map corresponding to the jump connection in the flux prediction module, and the corresponding feature map generated in the dose prediction module are concatenated and then convolved. This allows the model to capture continuous changes in organs, and ultimately generates a radiotherapy plan with higher dose distribution accuracy and higher radiotherapy quality.

Description

Technical Field

[0001] The present invention belongs to the field of artificial intelligence medical technology and relates to the generation of radiotherapy plans, and in particular to a radiotherapy plan generation method, system, device and medium based on ordinary differential equations. Background Art

[0002] Radiation therapy (RT) is an important tumor treatment method. Its main method is to provide targeted radiation to cancer cells while minimizing exposure to healthy organs. Therefore, the development of high-quality radiotherapy plans is crucial in the treatment of tumors. Over the past few decades, intensity-modulated radiation therapy (IMRT) has been widely adopted as a treatment target. IMRT uses adjustable leaves of a multi-leaf collimato (MLC) to adjust the shape and intensity of the radiation beam. In clinical practice, MLC sequences are represented as flux maps, so the flux map can be regarded as the radiotherapy plan itself. However, the traditional method of creating IMRT treatment plans, known as inverse optimization, is a time-consuming and experience-dependent process.

[0003] To reduce the time cost of radiotherapy planning and further mitigate the impact of human planning habits on treatment outcomes, various semi-automated treatment planning methods have been proposed in recent years. These methods can be roughly divided into two categories: knowledge-based planning (KBP) methods and deep learning-based dose prediction (DLDP) methods. The KBP method aims to predict the dose-volume histogram (DVH) for new patients based on historical data of patients with similar anatomical features, and use this as a reference to assist physicists in completing radiotherapy planning. The DLDP method, on the other hand, uses a deep neural network trained on clinical cases to predict the dose distribution of new patients.

[0004] In recent years, with the rapid development of artificial intelligence technology, the application of dose prediction methods based on deep learning has become more and more extensive due to its fast computing speed and high degree of automation.

[0005] The invention patent application with application number 202110142971.X discloses a method for automatically generating radiotherapy plans based on deep learning, which includes: step A, preprocessing the case database to obtain case feature images and case flux maps; step B, using the case feature images and case flux maps to train the flux map prediction neural network, and obtaining the optimal flux map prediction neural network through cross-validation; step C, reading and preprocessing the medical digital image storage and communication standard file of the radiotherapy patient to obtain the feature image of the patient to be radiotreated; step D, passing the feature image of the patient to be radiotreated to the prediction neural network to obtain the predicted flux map of the patient to be radiotreated, and sending the predicted flux map of the patient to be radiotreated to the planning system, thereby generating a predicted radiotherapy plan for the patient to be radiotreated. By introducing deep learning of convolutional neural networks to generate directly usable radiotherapy plans, the dependence on the personal experience of physicists is reduced, the error of manual prediction is reduced, the time for formulating radiotherapy plans is shortened, and personalized treatment plans are provided.

[0006] Application number 202310869407.7 also discloses a radiotherapy verification plan dose generation method, system, electronic device and storage medium, the method including: obtaining a case verification plan data set, and obtaining projection information of the planned radiation field through data analysis; using the phantom CT image data, phantom contour structure data, and projection information corresponding to the case verification plan data set as input data of a deep learning model, and the dose distribution map as the output of the deep learning model, and training the deep learning model until a dose calculation network model is obtained; using the dose calculation network model to perform dose calculation on the case information to be calculated, the dose of the patient verification plan can be quickly generated, thereby improving the accuracy of plan verification; automatically calculating the dose through the deep learning model reduces interference from human factors, reduces error rate, and improves work efficiency.

[0007] As in the aforementioned invention patent application, most existing technologies obtain projection maps and feature maps and input them into a network model to ultimately obtain a radiotherapy plan. However, in the process of generating a radiotherapy plan, in addition to considering the projection map, or the combination of the projection map and the flux map, it lacks an important reference basis - the dose map. As a result, when radiotherapy is subsequently performed according to the generated radiotherapy plan, the treatment quality is low and the radiotherapy effect is unsatisfactory. In addition, most existing network models use a single U-Net network, which cannot capture the continuous changes of organs. The dose distribution accuracy in the generated radiotherapy plan is low, and the radiotherapy quality is low. Summary of the Invention

[0008] The purpose of the present invention is to provide a method, system, device and medium for generating a radiotherapy plan based on ordinary differential equations in order to solve the technical problems of low dose distribution accuracy and poor radiotherapy quality in the radiotherapy plans generated in the prior art.

[0009] In order to achieve the above-mentioned purpose, the present invention specifically adopts the following technical solutions:

[0010] A method for generating a radiotherapy plan based on ordinary differential equations comprises the following steps:

[0011] Step 1, obtaining sample image data;

[0012] Obtain CT outlines, radiotherapy dose maps, and radiotherapy flux maps of the tumor site, and project the CT outlines onto a two-dimensional plane perpendicular to the direction of the radiotherapy rays to obtain a projection map;

[0013] Step 2: construct a radiotherapy plan generation model;

[0014] Constructing a radiotherapy plan generation model, which includes a dose prediction module and a flux prediction module;

[0015] In the dose prediction module, the projection image is downsampled and upsampled to output the dose map;

[0016] In the flux prediction module, the projection map and the dose map are downsampled, the projection map is spliced ​​after downsampling, and then the flux map is output after upsampling;

[0017] Step 3: training the radiotherapy plan generation model;

[0018] The sample image data obtained in step 1 is used to train the radiotherapy plan generation model constructed in step 2;

[0019] Step 4: Generate radiotherapy plan;

[0020] Obtain a CT outline of the tumor site for which a plan is to be generated, and project the CT outline onto a two-dimensional plane perpendicular to the direction of the radiotherapy ray to obtain a projection image to be tested; input the projection image to be tested into the radiotherapy plan generation model trained in step 3, and the radiotherapy plan generation model outputs a radiotherapy plan.

[0021] Further, in step 2, the dose prediction module includes a first dose convolution submodule, a first dose downsampling submodule, a second dose convolution submodule, a second dose downsampling submodule, a third dose convolution submodule, a third dose downsampling submodule, a fourth dose convolution submodule, a first dose upsampling submodule, a fifth dose convolution submodule, a second dose upsampling submodule, a sixth dose convolution submodule, a third dose upsampling submodule, and a seventh dose convolution submodule, which are arranged in sequence;

[0022] The projection image is obtained after the first dose convolution submodule, the first dose characteristic image is obtained after the first dose downsampling submodule, the second dose characteristic image is obtained after the second dose convolution submodule, the third dose characteristic image is obtained after the second dose downsampling submodule, the fourth dose characteristic image is obtained after the third dose convolution submodule, the fifth dose characteristic image is obtained after the third dose downsampling submodule, the sixth dose characteristic image is obtained after the fifth dose characteristic image is obtained after the third dose downsampling submodule, the seventh dose characteristic image is obtained after the seventh dose characteristic image is obtained after the first dose convolution submodule. After the dose upsampling submodule, the eighth dose characteristic map is obtained, the eighth dose characteristic map is spliced ​​with the fifth dose characteristic map and passed through the fifth dose convolution submodule to obtain the ninth dose characteristic map, the ninth dose characteristic map is passed through the second dose upsampling submodule to obtain the tenth dose characteristic map, the tenth dose characteristic map is spliced ​​with the third dose characteristic map and passed through the sixth dose convolution submodule to obtain the eleventh dose characteristic map, the eleventh dose characteristic map is passed through the third dose upsampling submodule to obtain the twelfth dose characteristic map, the twelfth dose characteristic map is spliced ​​with the first dose characteristic map and passed through the seventh dose convolution submodule to obtain the dose map.

[0023] Further, in step 2, the flux prediction module includes a first flux convolution submodule, a first flux downsampling submodule, a second flux convolution submodule, a second flux downsampling submodule, a third flux convolution submodule, a third flux downsampling submodule, a fourth flux convolution submodule, a first flux upsampling submodule, a fifth flux convolution submodule, a second flux upsampling submodule, a sixth flux convolution submodule, a third flux upsampling submodule, a seventh flux convolution submodule, an eighth flux convolution submodule, a fourth flux downsampling submodule, a ninth flux convolution submodule, a fifth flux downsampling submodule, a tenth flux convolution submodule, a sixth flux downsampling submodule, and an eleventh flux convolution submodule, which are arranged in sequence;

[0024] The projection image is passed through the eighth flux convolution submodule to obtain a thirteenth flux feature map, the thirteenth flux feature map is passed through the fourth flux downsampling submodule to obtain a fourteenth flux feature map, the fourteenth flux feature map is passed through the ninth flux convolution submodule to obtain a fifteenth flux feature map, the fifteenth flux feature map is passed through the fifth flux downsampling submodule to obtain a sixteenth flux feature map, the sixteenth flux feature map is passed through the tenth flux convolution submodule to obtain a seventeenth flux feature map, the seventeenth flux feature map is passed through the sixth flux downsampling submodule to obtain an eighteenth flux feature map, and the eighteenth flux feature map is passed through the eleventh flux convolution submodule to obtain a nineteenth flux feature map;

[0025] The projection map and the dose map output by the dose prediction module are passed through the first flux convolution submodule to obtain a first flux feature map, the first flux feature map is passed through the first flux downsampling submodule to obtain a second flux feature map, the second flux feature map is passed through the second flux convolution submodule to obtain a third flux feature map, the third flux feature map is passed through the second flux downsampling submodule to obtain a fourth flux feature map, the fourth flux feature map is passed through the third flux convolution submodule to obtain a fifth flux feature map, the fifth flux feature map is passed through the third flux downsampling submodule to obtain a sixth flux feature map, the sixth flux feature map is passed through the fourth flux convolution submodule to obtain a seventh flux feature map, the seventh flux feature map is added to the nineteenth flux feature map and passed through the first flux upsampling submodule to obtain the ninth flux feature map. Eight flux characteristic maps, the eighth flux characteristic map, the fifth flux characteristic map and the fifth dose characteristic map output by the dose prediction module are spliced ​​and passed through the fifth flux convolution submodule to obtain the ninth flux characteristic map, the ninth flux characteristic map is passed through the second flux upsampling submodule to obtain the tenth flux characteristic map, the tenth flux characteristic map, the third flux characteristic map and the third dose characteristic map output by the dose prediction module are spliced ​​and passed through the sixth flux convolution submodule to obtain the eleventh flux characteristic map, the eleventh flux characteristic map is passed through the third flux upsampling submodule to obtain the twelfth flux characteristic map, the twelfth flux characteristic map, the first flux characteristic map and the first dose characteristic map output by the dose prediction module are spliced ​​and passed through the seventh flux convolution submodule to obtain a flux map.

[0026] Furthermore, the dose prediction module and the flux prediction module are both provided with a NODE module and a 1*1 convolution layer. The feature map output by the seventh dose convolution submodule of the dose prediction module passes through the NODE module and the 1*1 convolution layer in sequence to output the dose map. The feature map output by the seventh flux convolution submodule of the flux prediction module passes through the NODE module and the 1*1 convolution layer in sequence to output the flux map.

[0027] Furthermore, the ordinary differential equation used by the NODE module is:

[0028]

[0029]

[0030] in, Indicates input, Represents the state value of the model, represents a learnable function, and Both represent learnable parameters.

[0031] Furthermore, the trained radiotherapy plan generation model was tested using the test set in the sample image data. The evaluation indicators in the test process included MAE, MSE, and PSNR.

[0032] The calculation method of MAE indicator is:

[0033]

[0034] The calculation method of MSE indicator is:

[0035]

[0036] The calculation method of PSNR indicator is:

[0037]

[0038] in, Indicates the pixels, Represents the total number of pixels, Indicates the first The intensity of the pixel, Indicates the first The intensity of the pixel, Indicates the maximum value of the pixel in the entire flux map.

[0039] A radiotherapy plan generation system based on ordinary differential equations, comprising:

[0040] The sample image data acquisition module is used to obtain the CT outline, radiotherapy dose map and radiotherapy flux map of the tumor site, and project the CT outline onto a two-dimensional plane perpendicular to the direction of the radiotherapy ray to obtain a projection map;

[0041] A radiotherapy plan generation model construction module is used to construct a radiotherapy plan generation model, which includes a dose prediction module and a flux prediction module;

[0042] In the dose prediction module, the projection image is downsampled and upsampled to output the dose map;

[0043] In the flux prediction module, the projection map and the dose map are downsampled, the projection map is spliced ​​after downsampling, and then the flux map is output after upsampling;

[0044] a radiotherapy plan generation model training module, configured to train the radiotherapy plan generation model constructed by the radiotherapy plan generation model construction module using the sample image data acquired by the sample image data acquisition module;

[0045] The radiotherapy plan generation module is used to obtain the CT outline of the tumor site for which the plan is to be generated, and project the CT outline onto a two-dimensional plane perpendicular to the direction of the radiotherapy ray to obtain a projection image to be tested; the projection image to be tested is input into the radiotherapy plan generation model trained by the radiotherapy plan generation model training module, and the radiotherapy plan generation model outputs the radiotherapy plan.

[0046] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the above method.

[0047] A computer-readable storage medium stores a computer program, which, when executed by a processor, causes the processor to perform the steps of the above method.

[0048] The beneficial effects of the present invention are as follows:

[0049] 1. In the present invention, a dose prediction module and a flux prediction module are set in the radiotherapy plan generation model, both of which adopt the U-Net network structure. The projection map is used as the input of the dose prediction module, and the output and projection map of the dose prediction module are used as the input of the flux prediction module. When the flux prediction module is upsampled, the feature map obtained after upsampling, the feature map corresponding to the jump connection in the flux prediction module, and the corresponding feature map generated in the dose prediction module are spliced ​​and then convolution processing is performed. In this way, the model can capture the continuous changes of the organ, and the dose distribution accuracy in the finally generated radiotherapy plan is higher, and the radiotherapy quality is higher.

[0050] 2. In the present invention, by introducing a neural network model based on neural ordinary differential equations, fully automated radiotherapy plan generation can be achieved, reducing the reliance on manual intervention in traditional methods, improving the efficiency and consistency of radiotherapy plan formulation, and reducing the workload of manual outlining and plan optimization.

[0051] 3. In the present invention, the characteristics of neural ordinary differential equations are utilized to better capture the continuous changes of tumor target areas and organs at risk in CT images, thereby generating more accurate dose distributions; compared with traditional radiotherapy planning methods based on static images, the present application can automatically take into account the complexity of the anatomical structure of individual patients and provide personalized treatment plans.

[0052] 4. In the present invention, the outlined image is projected using a ray tracing algorithm to convert three-dimensional data into a two-dimensional representation, thereby improving computational efficiency while retaining key anatomical information. This data preprocessing method can better simulate the projection effects during actual radiotherapy, making the plan generated by the model more in line with clinical needs.

[0053] 5. In the present invention, the time for plan generation is significantly reduced through a fully automated process, so that radiotherapy plans can be completed in a shorter time, thereby providing treatment to patients more quickly.

[0054] 6. In the present invention, through the training and evaluation process of the model, the dose limitation of organs at risk can be automatically taken into account when generating radiotherapy plans, thereby maximizing the protection of healthy tissues, optimizing the treatment dose of tumors, and improving the safety and effectiveness of treatment.

[0055] 7. The fully automated method of the present invention can reduce the dependence on the operator's experience, reduce the impact of human factors on the quality of radiotherapy plans, and make the formulation of radiotherapy plans more standardized and normalized.

[0056] 8. In the present invention, by introducing the NODE module with excellent ability in processing complex and nonlinear features into the two prediction modules, the NODE module can capture complex dynamic behaviors, significantly enhance its expression ability in processing nonlinear systems, and improve the accuracy of dose distribution recognition. The dose distribution in the final generated radiotherapy plan is more accurate and the radiotherapy quality is higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a schematic flow diagram of the present invention;

[0058] Figure 2 is a schematic diagram of a radiotherapy plan generation model in the present invention;

[0059] Among them, splicing means merging two feature maps into one feature map along a specified dimension, so that its feature information is expanded in this dimension, thereby retaining and integrating all the information of the two sets of features; multiplication means performing element-by-element multiplication operations on the corresponding position elements of the two feature maps; addition means performing element-by-element addition operations on the corresponding position elements of the two feature maps, thereby fusing the two sets of feature information and realizing the superposition and comprehensive expression of features;

[0060] Figure 3 is a schematic diagram of a dose prediction module in the radiotherapy plan generation model of the present invention;

[0061] Figure 4 is a schematic diagram of a flux prediction module in the radiotherapy plan generation model of the present invention;

[0062] Figure 5 It is a schematic diagram of projection of target area organs at risk in the present invention. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments.

[0064] Therefore, based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of the present invention.

[0065] Example 1

[0066] This embodiment provides a radiotherapy plan generation method based on ordinary differential equations, which is used to automatically generate a radiotherapy plan with higher dose distribution accuracy and higher radiotherapy quality. Figure 1 As shown, it includes the following steps:

[0067] Step 1, obtaining sample image data;

[0068] Obtain CT outlines, radiotherapy dose maps, and radiotherapy flux maps of the tumor site, and project the CT outlines onto a two-dimensional plane perpendicular to the direction of the radiotherapy rays to obtain a projection map.

[0069] Real case data were collected from West China Hospital of Sichuan University as sample data, including patients' CT outlines, radiotherapy dose maps, and radiotherapy flux maps, and the above data were organized according to the structure.

[0070] Acquisition of CT data: A physicist obtains CT image data by using a CT machine to scan the patient's tumor area. One scan will generate multiple continuous CT images.

[0071] Organ delineation data acquisition: The physicist delineates the target area and organs at risk on the acquired CT images based on the pathological characteristics of the case.

[0072] Flux map data acquisition: The physicist uses the radiotherapy planning system to make the radiotherapy plan based on the pathological outline structure.

[0073] Dose data acquisition: The physicist simulates the radiotherapy process of the prepared radiotherapy plan to obtain dose data.

[0074] All types of data are organized with case as the primary key.

[0075] In order to improve the efficiency of model training and the accuracy of plan generation, the sample data needs to be processed to make it more suitable for the input format of the neural network. The specific data processing process is as follows:

[0076] like Figure 5 As shown in the figure, a ray tracing algorithm is first used to project the 3D outlines of the target area and organs at risk in the CT image onto a 2D plane perpendicular to the direction of the radiotherapy rays, resulting in a projection image. This projection simulates the actual projection effect during radiotherapy, simplifying the processing of 3D images while preserving important structural information.

[0077] The projection images are then normalized to ensure that all data are of the same scale. This step can reduce the model's sensitivity to data segmentation, allowing it to focus more on feature extraction of target areas and organs at risk.

[0078] Finally, the standardized dataset was divided into training, validation, and test sets in a ratio of 8:1:1. The training set was used to train the model and optimize its parameters; the validation set was used to evaluate and adjust hyperparameters; and the test set was used for final model evaluation to verify its performance in real-world applications.

[0079] Step 2: construct a radiotherapy plan generation model;

[0080] Constructing a radiotherapy plan generation model, which includes a dose prediction module and a flux prediction module;

[0081] In the dose prediction module, the projection image is downsampled and upsampled to output the dose map;

[0082] In the flux prediction module, the projection map and the dose map are downsampled, the projection map is spliced ​​after downsampling, and then the flux map is output after upsampling;

[0083] The specific working process of the radiotherapy plan generation model, such as Figure 2 shown.

[0084] like Figure 3 As shown, the dose prediction module adopts a U-Net network structure, which includes a first dose convolution submodule, a first dose downsampling submodule, a second dose convolution submodule, a second dose downsampling submodule, a third dose convolution submodule, a third dose downsampling submodule, a fourth dose convolution submodule, a first dose upsampling submodule, a fifth dose convolution submodule, a second dose upsampling submodule, a sixth dose convolution submodule, a third dose upsampling submodule, a seventh dose convolution submodule, a NODE module, and a 1*1 convolution layer, which are arranged in sequence.

[0085] The projection image is obtained after the first dose convolution submodule, the first dose characteristic map is obtained after the first dose downsampling submodule, the second dose characteristic map is obtained after the second dose convolution submodule, the third dose characteristic map is obtained after the second dose downsampling submodule, the fourth dose characteristic map is obtained after the third dose convolution submodule, the fifth dose characteristic map is obtained after the third dose downsampling submodule, the sixth dose characteristic map is obtained after the fifth dose characteristic map is obtained after the third dose downsampling submodule, the seventh dose characteristic map is obtained after the fourth dose convolution submodule, the eighth dose characteristic map is obtained after the first dose upsampling submodule, and the eighth dose characteristic map is spliced ​​with the fifth dose characteristic map (that is, the fifth dose characteristic map is spliced ​​with the eighth dose characteristic map by a jump connection). The dose characteristic map is connected and spliced) and passes through the fifth dose convolution submodule to obtain the ninth dose characteristic map, the ninth dose characteristic map is passed through the second dose upsampling submodule to obtain the tenth dose characteristic map, the tenth dose characteristic map is spliced ​​with the third dose characteristic map (that is, the third dose characteristic map is connected and spliced ​​with the tenth dose characteristic map through a jump connection) and passes through the sixth dose convolution submodule to obtain the eleventh dose characteristic map, the eleventh dose characteristic map is passed through the third dose upsampling submodule to obtain the twelfth dose characteristic map, the twelfth dose characteristic map is spliced ​​with the first dose characteristic map (that is, the first dose characteristic map is connected and spliced ​​with the twelfth dose characteristic map through a jump connection) and passes through the seventh dose convolution submodule, the NODE module, and the 1*1 convolution layer to obtain the dose map.

[0086] Among them, the first dose convolution submodule, the second dose convolution submodule, the third dose convolution submodule, the fourth dose convolution submodule, the fifth dose convolution submodule and the sixth dose convolution submodule all include two layers of 3*3 convolution layers, and the seventh dose convolution submodule includes one layer of 3*3 convolution layer.

[0087] In the dose prediction module, the projection map undergoes three rounds of encoding, gradually reducing the projection map size from 192*192 to a 24*24 feature map, while increasing the number of channels to 1024. This encoding process focuses on capturing abstract latent features of different scales; subsequently, the final output dose map is generated through three decoding stages. Each decoding stage includes upsampling and convolution operations to reconstruct the dose map based on the encoded features, and jump connections help retrieve features of the same size from the encoder, thereby facilitating the reconstruction process; finally, the NODE module is used for nonlinear remapping to obtain the predicted dose map.

[0088] like Figure 4As shown, the flux prediction module also adopts the U-Net network structure, which includes a first flux convolution submodule, a first flux downsampling submodule, a second flux convolution submodule, a second flux downsampling submodule, a third flux convolution submodule, a third flux downsampling submodule, a fourth flux convolution submodule, a first flux upsampling submodule, a fifth flux convolution submodule, a second flux upsampling submodule, a sixth flux convolution submodule, a third flux upsampling submodule, a seventh flux convolution submodule, an eighth flux convolution submodule, a fourth flux downsampling submodule, a ninth flux convolution submodule, a fifth flux downsampling submodule, a tenth flux convolution submodule, a sixth flux downsampling submodule, and an eleventh flux convolution submodule, which are arranged in sequence.

[0089] The flux prediction module has two inputs. One is to input the projection map into the eighth flux convolution submodule of the flux prediction module. The other is to splice the projection map and the dose map output by the dose prediction module and then input them into the first flux convolution submodule of the flux prediction module.

[0090] The projection image obtains the thirteenth flux feature map after passing through the eighth flux convolution submodule, the thirteenth flux feature map obtains the fourteenth flux feature map after passing through the fourth flux downsampling submodule, the fourteenth flux feature map obtains the fifteenth flux feature map after passing through the ninth flux convolution submodule, the fifteenth flux feature map obtains the sixteenth flux feature map after passing through the fifth flux downsampling submodule, the sixteenth flux feature map obtains the seventeenth flux feature map after passing through the tenth flux convolution submodule, the seventeenth flux feature map obtains the eighteenth flux feature map after passing through the sixth flux downsampling submodule, and the eighteenth flux feature map obtains the nineteenth flux feature map after passing through the eleventh flux convolution submodule.

[0091] The projection image and the dose image output by the dose prediction module (that is, the two images are spliced ​​and input) are passed through the first flux convolution submodule to obtain the first flux characteristic image, the first flux characteristic image is passed through the first flux downsampling submodule to obtain the second flux characteristic image, the second flux characteristic image is passed through the second flux convolution submodule to obtain the third flux characteristic image, the third flux characteristic image is passed through the second flux downsampling submodule to obtain the fourth flux characteristic image, the fourth flux characteristic image is passed through the third flux convolution submodule to obtain the fifth flux characteristic image, the fifth flux characteristic image is passed through the third flux downsampling submodule to obtain the sixth flux characteristic image, the sixth flux characteristic image is passed through the fourth flux convolution submodule to obtain the seventh flux characteristic image, the seventh flux characteristic image is added to the nineteenth flux characteristic image and passed through the first flux upsampling submodule to obtain the eighth flux characteristic image, the eighth flux characteristic image, the fifth flux characteristic image and the fifth dose characteristic image output by the dose prediction module are spliced ​​(that is, the fifth flux characteristic image, the fifth dose characteristic image). The ninth flux characteristic map is obtained after passing through the fifth flux convolution submodule. The ninth flux characteristic map is obtained after passing through the second flux upsampling submodule. The tenth flux characteristic map, the third flux characteristic map and the third dose characteristic map outputted midway by the dose prediction module are spliced ​​(i.e., the third flux characteristic map and the third dose characteristic map are connected and spliced ​​with the tenth flux characteristic map by means of a jump connection) and are obtained after passing through the sixth flux convolution submodule. The eleventh flux characteristic map is obtained after passing through the third flux upsampling submodule. The twelfth flux characteristic map, the first flux characteristic map and the first dose characteristic map outputted midway by the dose prediction module are spliced ​​(i.e., the first flux characteristic map and the first dose characteristic map are connected and spliced ​​with the twelfth flux characteristic map by means of a jump connection) and are obtained after passing through the seventh flux convolution submodule.

[0092] The structure of the flux prediction module is consistent with that of the dose prediction module. The difference is that it takes the dose map predicted by the dose prediction module and the projection map of the target area's organs at risk as input at the same time. After a series of encoding and decoding operations, the feature map obtained is sent to the NODE module for nonlinear remapping to obtain the predicted flux map, that is, the radiotherapy plan.

[0093] The ordinary differential equation used by the NODE module is:

[0094]

[0095]

[0096] in, Indicates input, Represents the state value of the model, represents a learnable function, and Both represent learnable parameters.

[0097] Step 3: training the radiotherapy plan generation model;

[0098] The sample image data obtained in step 1 is used to train the radiotherapy plan generation model constructed in step 2.

[0099] The training set is input into the radiotherapy plan generation model, and forward calculation and back propagation are performed in sequence, repeated multiple times until the model converges. Training can be stopped when the output on the validation set is the minimum value among multiple trainings, thereby obtaining the optimal model.

[0100] The model can be trained using existing technologies, and the parameter settings, loss functions, etc. in the training process can be directly applied to existing technologies without the need for creative work.

[0101] After training and verifying the model, the model needs to be tested. Use the test set to test the model. The evaluation indicators in the test process can be selected from the three indicators of MAE, MSE, and PSNR.

[0102] The calculation method of MAE indicator is:

[0103]

[0104] The calculation method of MSE indicator is:

[0105]

[0106] The calculation method of PSNR indicator is:

[0107]

[0108] in, Indicates the pixels, represents the total number of pixels, Indicates the first The intensity of the pixel, Indicates the first The intensity of the pixel, Indicates the maximum value of the pixel in the entire flux map.

[0109] Step 4: Generate radiotherapy plan;

[0110] Obtain a CT outline of the tumor site for which a plan is to be generated, and project the CT outline onto a two-dimensional plane perpendicular to the direction of the radiotherapy ray to obtain a projection image to be tested; input the projection image to be tested into the radiotherapy plan generation model trained in step 3, and the radiotherapy plan generation model outputs a radiotherapy plan.

[0111] Example 2

[0112] This embodiment provides a radiotherapy plan generation system based on ordinary differential equations, which is used to automatically generate radiotherapy plans with higher dose distribution accuracy and higher radiotherapy quality. It includes:

[0113] The sample image data acquisition module is used to obtain the CT outline, radiotherapy dose map and radiotherapy flux map of the tumor site, and project the CT outline onto a two-dimensional plane perpendicular to the direction of the radiotherapy ray to obtain a projection map.

[0114] Real case data were collected from West China Hospital of Sichuan University as sample data, including patients' CT outlines, radiotherapy dose maps, and radiotherapy flux maps, and the above data were organized according to the structure.

[0115] Acquisition of CT data: A physicist obtains CT image data by scanning the patient's tumor area using a CT machine. One scan will generate multiple continuous CT images.

[0116] Organ delineation data acquisition: The physicist delineates the target area and organs at risk on the acquired CT images based on the pathological characteristics of the case.

[0117] Flux map data acquisition: The physicist uses the radiotherapy planning system to make the radiotherapy plan based on the outlined structure of the pathology.

[0118] Dose data acquisition: The physicist simulates the radiotherapy process of the prepared radiotherapy plan to obtain dose data.

[0119] All types of data are organized with case as the primary key.

[0120] In order to improve the efficiency of model training and the accuracy of plan generation, the sample data needs to be processed to make it more suitable for the input format of the neural network. The specific data processing process is as follows:

[0121] like Figure 5 As shown in the figure, a ray tracing algorithm is first used to project the 3D outlines of the target area and organs at risk in the CT image onto a 2D plane perpendicular to the direction of the radiotherapy rays, resulting in a projection image. This projection simulates the actual projection effect during radiotherapy, simplifying the processing of 3D images while preserving important structural information.

[0122] The projection images are then normalized to ensure that all data are of the same scale. This step can reduce the model's sensitivity to data segmentation, allowing it to focus more on feature extraction of target areas and organs at risk.

[0123] Finally, the standardized dataset was divided into training, validation, and test sets in a ratio of 8:1:1. The training set was used to train the model and optimize its parameters; the validation set was used to evaluate and adjust hyperparameters; and the test set was used for final model evaluation to verify its performance in real-world applications.

[0124] A radiotherapy plan generation model construction module is used to construct a radiotherapy plan generation model, which includes a dose prediction module and a flux prediction module;

[0125] In the dose prediction module, the projection image is downsampled and upsampled to output the dose map;

[0126] In the flux prediction module, the projection map and the dose map are downsampled, the projection map is spliced ​​after downsampling, and then the flux map is output after upsampling;

[0127] The specific working process of the radiotherapy plan generation model, such as Figure 2 shown.

[0128] like Figure 3 As shown, the dose prediction module adopts a U-Net network structure, which includes a first dose convolution submodule, a first dose downsampling submodule, a second dose convolution submodule, a second dose downsampling submodule, a third dose convolution submodule, a third dose downsampling submodule, a fourth dose convolution submodule, a first dose upsampling submodule, a fifth dose convolution submodule, a second dose upsampling submodule, a sixth dose convolution submodule, a third dose upsampling submodule, a seventh dose convolution submodule, a NODE module, and a 1*1 convolution layer, which are arranged in sequence.

[0129] The projection image is obtained after the first dose convolution submodule, the first dose characteristic map is obtained after the first dose downsampling submodule, the second dose characteristic map is obtained after the second dose convolution submodule, the third dose characteristic map is obtained after the second dose downsampling submodule, the fourth dose characteristic map is obtained after the third dose convolution submodule, the fifth dose characteristic map is obtained after the third dose downsampling submodule, the sixth dose characteristic map is obtained after the fifth dose characteristic map is obtained after the third dose downsampling submodule, the seventh dose characteristic map is obtained after the fourth dose convolution submodule, the eighth dose characteristic map is obtained after the first dose upsampling submodule, and the eighth dose characteristic map is spliced ​​with the fifth dose characteristic map (that is, the fifth dose characteristic map is spliced ​​with the eighth dose characteristic map by a jump connection). The dose characteristic map is connected and spliced) and passes through the fifth dose convolution submodule to obtain the ninth dose characteristic map, the ninth dose characteristic map is passed through the second dose upsampling submodule to obtain the tenth dose characteristic map, the tenth dose characteristic map is spliced ​​with the third dose characteristic map (that is, the third dose characteristic map is connected and spliced ​​with the tenth dose characteristic map through a jump connection) and passes through the sixth dose convolution submodule to obtain the eleventh dose characteristic map, the eleventh dose characteristic map is passed through the third dose upsampling submodule to obtain the twelfth dose characteristic map, the twelfth dose characteristic map is spliced ​​with the first dose characteristic map (that is, the first dose characteristic map is connected and spliced ​​with the twelfth dose characteristic map through a jump connection) and passes through the seventh dose convolution submodule, the NODE module, and the 1*1 convolution layer to obtain the dose map.

[0130] Among them, the first dose convolution submodule, the second dose convolution submodule, the third dose convolution submodule, the fourth dose convolution submodule, the fifth dose convolution submodule and the sixth dose convolution submodule all include two layers of 3*3 convolution layers, and the seventh dose convolution submodule includes one layer of 3*3 convolution layer.

[0131] In the dose prediction module, the projection map undergoes three rounds of encoding, gradually reducing the projection map size from 192*192 to a 24*24 feature map, while increasing the number of channels to 1024. This encoding process focuses on capturing abstract latent features of different scales; subsequently, the final output dose map is generated through three decoding stages. Each decoding stage includes upsampling and convolution operations to reconstruct the dose map based on the encoded features, and jump connections help retrieve features of the same size from the encoder, thereby facilitating the reconstruction process; finally, the NODE module is used for nonlinear remapping to obtain the predicted dose map.

[0132] like Figure 4As shown, the flux prediction module also adopts the U-Net network structure, which includes a first flux convolution submodule, a first flux downsampling submodule, a second flux convolution submodule, a second flux downsampling submodule, a third flux convolution submodule, a third flux downsampling submodule, a fourth flux convolution submodule, a first flux upsampling submodule, a fifth flux convolution submodule, a second flux upsampling submodule, a sixth flux convolution submodule, a third flux upsampling submodule, a seventh flux convolution submodule, an eighth flux convolution submodule, a fourth flux downsampling submodule, a ninth flux convolution submodule, a fifth flux downsampling submodule, a tenth flux convolution submodule, a sixth flux downsampling submodule, and an eleventh flux convolution submodule, which are arranged in sequence.

[0133] The flux prediction module has two inputs. One is to input the projection map into the eighth flux convolution submodule of the flux prediction module. The other is to splice the projection map and the dose map output by the dose prediction module and then input them into the first flux convolution submodule of the flux prediction module.

[0134] The projection image obtains the thirteenth flux feature map after passing through the eighth flux convolution submodule, the thirteenth flux feature map obtains the fourteenth flux feature map after passing through the fourth flux downsampling submodule, the fourteenth flux feature map obtains the fifteenth flux feature map after passing through the ninth flux convolution submodule, the fifteenth flux feature map obtains the sixteenth flux feature map after passing through the fifth flux downsampling submodule, the sixteenth flux feature map obtains the seventeenth flux feature map after passing through the tenth flux convolution submodule, the seventeenth flux feature map obtains the eighteenth flux feature map after passing through the sixth flux downsampling submodule, and the eighteenth flux feature map obtains the nineteenth flux feature map after passing through the eleventh flux convolution submodule.

[0135] The projection image and the dose image output by the dose prediction module (that is, the two images are spliced ​​and input) are passed through the first flux convolution submodule to obtain the first flux characteristic image, the first flux characteristic image is passed through the first flux downsampling submodule to obtain the second flux characteristic image, the second flux characteristic image is passed through the second flux convolution submodule to obtain the third flux characteristic image, the third flux characteristic image is passed through the second flux downsampling submodule to obtain the fourth flux characteristic image, the fourth flux characteristic image is passed through the third flux convolution submodule to obtain the fifth flux characteristic image, the fifth flux characteristic image is passed through the third flux downsampling submodule to obtain the sixth flux characteristic image, the sixth flux characteristic image is passed through the fourth flux convolution submodule to obtain the seventh flux characteristic image, the seventh flux characteristic image is added to the nineteenth flux characteristic image and passed through the first flux upsampling submodule to obtain the eighth flux characteristic image, the eighth flux characteristic image, the fifth flux characteristic image and the fifth dose characteristic image output by the dose prediction module are spliced ​​(that is, the fifth flux characteristic image, the fifth dose characteristic image). The ninth flux characteristic map is obtained after passing through the fifth flux convolution submodule. The ninth flux characteristic map is obtained after passing through the second flux upsampling submodule. The tenth flux characteristic map, the third flux characteristic map and the third dose characteristic map outputted midway by the dose prediction module are spliced ​​(i.e., the third flux characteristic map and the third dose characteristic map are connected and spliced ​​with the tenth flux characteristic map by means of a jump connection) and are obtained after passing through the sixth flux convolution submodule. The eleventh flux characteristic map is obtained after passing through the third flux upsampling submodule. The twelfth flux characteristic map, the first flux characteristic map and the first dose characteristic map outputted midway by the dose prediction module are spliced ​​(i.e., the first flux characteristic map and the first dose characteristic map are connected and spliced ​​with the twelfth flux characteristic map by means of a jump connection) and are obtained after passing through the seventh flux convolution submodule.

[0136] The structure of the flux prediction module is consistent with that of the dose prediction module. The difference is that it takes the dose map predicted by the dose prediction module and the projection map of the target area's organs at risk as input at the same time. After a series of encoding and decoding operations, the feature map obtained is sent to the NODE module for nonlinear remapping to obtain the predicted flux map, that is, the radiotherapy plan.

[0137] The ordinary differential equation used by the NODE module is:

[0138]

[0139]

[0140] in, Indicates input, Represents the state value of the model, represents a learnable function, and Both represent learnable parameters.

[0141] The radiotherapy plan generation model training module is used to train the radiotherapy plan generation model constructed by the radiotherapy plan generation model construction module using the sample image data acquired by the sample image data acquisition module.

[0142] The training set is input into the radiotherapy plan generation model, and forward calculation and back propagation are performed in sequence, repeated multiple times until the model converges. Training can be stopped when the output on the validation set is the minimum value among multiple trainings, thereby obtaining the optimal model.

[0143] The model can be trained using existing technologies, and the parameter settings, loss functions, etc. in the training process can be directly applied to existing technologies without the need for creative work.

[0144] The radiotherapy plan generation module is used to obtain the CT outline of the tumor site for which the plan is to be generated, and project the CT outline onto a two-dimensional plane perpendicular to the direction of the radiotherapy ray to obtain a projection image to be tested; the projection image to be tested is input into the radiotherapy plan generation model trained by the radiotherapy plan generation model training module, and the radiotherapy plan generation model outputs the radiotherapy plan.

[0145] Example 3

[0146] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of a radiotherapy plan generation method based on ordinary differential equations.

[0147] The computer device may be a desktop computer, a notebook computer, a PDA, a cloud server, etc. The computer device may interact with the user via a keyboard, a mouse, a remote control, a touchpad, or a voice control device.

[0148] The memory includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or D-interface display memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory may be an internal storage unit of the computer device, such as the hard disk or internal memory of the computer device. In other embodiments, the memory may also be an external storage device of the computer device, such as a plug-in hard disk equipped with the computer device, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Of course, the memory may also include both the internal storage unit of the computer device and its external storage device. In this embodiment, the memory is often used to store the operating system and various application software installed on the computer device, such as the program code of the ordinary differential equation-based radiotherapy plan generation method. In addition, the memory may also be used to temporarily store various types of data that have been output or are about to be output.

[0149] In some embodiments, the processor may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. The processor is typically used to control the overall operation of the computer device. In this embodiment, the processor is used to execute program code stored in the memory or process data, such as executing the program code for the ordinary differential equation-based radiotherapy plan generation method.

[0150] Example 4

[0151] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the processor is caused to perform the steps of a method for generating a radiotherapy plan based on ordinary differential equations.

[0152] The computer-readable storage medium stores an interface display program, and the interface display program can be executed by at least one processor to enable the at least one processor to perform the steps of the above-mentioned radiotherapy plan generation method based on ordinary differential equations.

[0153] Through the description of the above embodiments, those skilled in the art can clearly understand that the above embodiment method can be implemented by means of software plus the necessary general hardware platform, or of course by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server or network device, etc.) to execute the radiotherapy plan generation method based on ordinary differential equations described in the embodiment of the present application.

Claims

1. A method for generating a radiotherapy plan based on ordinary differential equations, characterized in that: The steps include: Step 1, obtaining sample image data; Obtain CT outlines, radiotherapy dose maps, and radiotherapy flux maps of the tumor site, and project the CT outlines onto a two-dimensional plane perpendicular to the direction of the radiotherapy rays to obtain a projection map; Step 2: construct a radiotherapy plan generation model; Constructing a radiotherapy plan generation model, which includes a dose prediction module and a flux prediction module; In the dose prediction module, the projection image is downsampled and upsampled to output the dose map; In the flux prediction module, the projection map and the dose map are downsampled, the projection map is spliced ​​after downsampling, and then the flux map is output after upsampling; Step 3: training the radiotherapy plan generation model; The sample image data obtained in step 1 is used to train the radiotherapy plan generation model constructed in step 2; Step 4: Generate radiotherapy plan; Obtain a CT outline of the tumor site for which a plan is to be generated, and project the CT outline onto a two-dimensional plane perpendicular to the direction of the radiotherapy rays to obtain a projection image to be tested; input the projection image to be tested into the radiotherapy plan generation model trained in step 3, and the radiotherapy plan generation model outputs a radiotherapy plan; In step 2, the dose prediction module includes a first dose convolution submodule, a first dose downsampling submodule, a second dose convolution submodule, a second dose downsampling submodule, a third dose convolution submodule, a third dose downsampling submodule, a fourth dose convolution submodule, a first dose upsampling submodule, a fifth dose convolution submodule, a second dose upsampling submodule, a sixth dose convolution submodule, a third dose upsampling submodule, and a seventh dose convolution submodule, which are arranged in sequence; The projection image is passed through the first dose convolution submodule to obtain the first dose characteristic map, the first dose characteristic map is passed through the first dose downsampling submodule to obtain the second dose characteristic map, the second dose characteristic map is passed through the second dose convolution submodule to obtain the third dose characteristic map, the third dose characteristic map is passed through the second dose downsampling submodule to obtain the fourth dose characteristic map, the fourth dose characteristic map is passed through the third dose convolution submodule to obtain the fifth dose characteristic map, the fifth dose characteristic map is passed through the third dose downsampling submodule to obtain the sixth dose characteristic map, the sixth dose characteristic map is passed through the fourth dose convolution submodule to obtain the seventh dose characteristic map, the seventh dose characteristic map is passed through the fourth dose convolution submodule. After passing through the first dose upsampling submodule, an eighth dose characteristic map is obtained, the eighth dose characteristic map is spliced ​​with the fifth dose characteristic map and passed through the fifth dose convolution submodule to obtain a ninth dose characteristic map, the ninth dose characteristic map is passed through the second dose upsampling submodule to obtain a tenth dose characteristic map, the tenth dose characteristic map is spliced ​​with the third dose characteristic map and passed through the sixth dose convolution submodule to obtain an eleventh dose characteristic map, the eleventh dose characteristic map is passed through the third dose upsampling submodule to obtain a twelfth dose characteristic map, the twelfth dose characteristic map is spliced ​​with the first dose characteristic map and passed through the seventh dose convolution submodule to obtain a dose map; In step 2, the flux prediction module includes a first flux convolution submodule, a first flux downsampling submodule, a second flux convolution submodule, a second flux downsampling submodule, a third flux convolution submodule, a third flux downsampling submodule, a fourth flux convolution submodule, a first flux upsampling submodule, a fifth flux convolution submodule, a second flux upsampling submodule, a sixth flux convolution submodule, a third flux upsampling submodule, a seventh flux convolution submodule, an eighth flux convolution submodule, a fourth flux downsampling submodule, a ninth flux convolution submodule, a fifth flux downsampling submodule, a tenth flux convolution submodule, a sixth flux downsampling submodule, and an eleventh flux convolution submodule, which are arranged in sequence; The projection image is passed through the eighth flux convolution submodule to obtain a thirteenth flux feature map, the thirteenth flux feature map is passed through the fourth flux downsampling submodule to obtain a fourteenth flux feature map, the fourteenth flux feature map is passed through the ninth flux convolution submodule to obtain a fifteenth flux feature map, the fifteenth flux feature map is passed through the fifth flux downsampling submodule to obtain a sixteenth flux feature map, the sixteenth flux feature map is passed through the tenth flux convolution submodule to obtain a seventeenth flux feature map, the seventeenth flux feature map is passed through the sixth flux downsampling submodule to obtain an eighteenth flux feature map, and the eighteenth flux feature map is passed through the eleventh flux convolution submodule to obtain a nineteenth flux feature map; The projection map and the dose map output by the dose prediction module are passed through the first flux convolution submodule to obtain a first flux feature map, the first flux feature map is passed through the first flux downsampling submodule to obtain a second flux feature map, the second flux feature map is passed through the second flux convolution submodule to obtain a third flux feature map, the third flux feature map is passed through the second flux downsampling submodule to obtain a fourth flux feature map, the fourth flux feature map is passed through the third flux convolution submodule to obtain a fifth flux feature map, the fifth flux feature map is passed through the third flux downsampling submodule to obtain a sixth flux feature map, the sixth flux feature map is passed through the fourth flux convolution submodule to obtain a seventh flux feature map, the seventh flux feature map is added to the nineteenth flux feature map and passed through the first flux upsampling submodule to obtain the ninth flux feature map. The eighth flux characteristic map, the eighth flux characteristic map, the fifth flux characteristic map and the eighth dose characteristic map outputted midway by the dose prediction module are spliced ​​and passed through the fifth flux convolution submodule to obtain the ninth flux characteristic map. The ninth flux characteristic map is passed through the second flux upsampling submodule to obtain the tenth flux characteristic map. The tenth flux characteristic map, the third flux characteristic map and the tenth dose characteristic map outputted midway by the dose prediction module are spliced ​​and passed through the sixth flux convolution submodule to obtain the eleventh flux characteristic map. The eleventh flux characteristic map is passed through the third flux upsampling submodule to obtain the twelfth flux characteristic map. The twelfth flux characteristic map, the first flux characteristic map and the twelfth dose characteristic map outputted midway by the dose prediction module are spliced ​​and passed through the seventh flux convolution submodule to obtain a flux map.

2. The method for generating a radiotherapy plan based on ordinary differential equations according to claim 1, wherein: Both the dose prediction module and the flux prediction module are equipped with a NODE module and a 1*1 convolution layer. The feature map output by the seventh dose convolution submodule of the dose prediction module passes through the NODE module and the 1*1 convolution layer in sequence to output the dose map. The feature map output by the seventh flux convolution submodule of the flux prediction module passes through the NODE module and the 1*1 convolution layer in sequence to output the flux map.

3. The method for generating a radiotherapy plan based on ordinary differential equations according to claim 2, wherein: The ordinary differential equation used by the NODE module is: in, Indicates input, Represents the state value of the model, represents a learnable function, and Both represent learnable parameters.

4. The method for generating a radiotherapy plan based on ordinary differential equations according to claim 1, wherein: The trained radiotherapy plan generation model was tested using the test set in the sample image data. The evaluation indicators during the test included MAE, MSE, and PSNR. The calculation method of MAE indicator is: The calculation method of MSE indicator is: The calculation method of PSNR indicator is: in, Indicates the pixels, Represents the total number of pixels, Indicates the first The intensity of the pixel, Indicates the first The intensity of the pixel, Indicates the maximum value of the pixel in the entire flux map.

5. A radiotherapy plan generation system based on ordinary differential equations, characterized in that: include: The sample image data acquisition module is used to obtain the CT outline, radiotherapy dose map and radiotherapy flux map of the tumor site, and project the CT outline onto a two-dimensional plane perpendicular to the direction of the radiotherapy ray to obtain a projection map; A radiotherapy plan generation model construction module is used to construct a radiotherapy plan generation model, which includes a dose prediction module and a flux prediction module; In the dose prediction module, the projection image is downsampled and upsampled to output the dose map; In the flux prediction module, the projection map and the dose map are downsampled, the projection map is spliced ​​after downsampling, and then the flux map is output after upsampling; a radiotherapy plan generation model training module, configured to train the radiotherapy plan generation model constructed by the radiotherapy plan generation model construction module using the sample image data acquired by the sample image data acquisition module; The radiotherapy plan generation module is used to obtain a CT outline of the tumor site for which a plan is to be generated, and project the CT outline onto a two-dimensional plane perpendicular to the direction of the radiotherapy ray to obtain a projection image to be measured; Inputting the projection image to be tested into the radiotherapy plan generation model trained by the radiotherapy plan generation model training module, and the radiotherapy plan generation model outputs the radiotherapy plan; In the radiotherapy plan generation model construction module, the dose prediction module includes a first dose convolution submodule, a first dose downsampling submodule, a second dose convolution submodule, a second dose downsampling submodule, a third dose convolution submodule, a third dose downsampling submodule, a fourth dose convolution submodule, a first dose upsampling submodule, a fifth dose convolution submodule, a second dose upsampling submodule, a sixth dose convolution submodule, a third dose upsampling submodule, and a seventh dose convolution submodule, which are arranged in sequence; The projection image is passed through the first dose convolution submodule to obtain the first dose characteristic map, the first dose characteristic map is passed through the first dose downsampling submodule to obtain the second dose characteristic map, the second dose characteristic map is passed through the second dose convolution submodule to obtain the third dose characteristic map, the third dose characteristic map is passed through the second dose downsampling submodule to obtain the fourth dose characteristic map, the fourth dose characteristic map is passed through the third dose convolution submodule to obtain the fifth dose characteristic map, the fifth dose characteristic map is passed through the third dose downsampling submodule to obtain the sixth dose characteristic map, the sixth dose characteristic map is passed through the fourth dose convolution submodule to obtain the seventh dose characteristic map, the seventh dose characteristic map is passed through the fourth dose convolution submodule. After passing through the first dose upsampling submodule, an eighth dose characteristic map is obtained, the eighth dose characteristic map is spliced ​​with the fifth dose characteristic map and passed through the fifth dose convolution submodule to obtain a ninth dose characteristic map, the ninth dose characteristic map is passed through the second dose upsampling submodule to obtain a tenth dose characteristic map, the tenth dose characteristic map is spliced ​​with the third dose characteristic map and passed through the sixth dose convolution submodule to obtain an eleventh dose characteristic map, the eleventh dose characteristic map is passed through the third dose upsampling submodule to obtain a twelfth dose characteristic map, the twelfth dose characteristic map is spliced ​​with the first dose characteristic map and passed through the seventh dose convolution submodule to obtain a dose map; In step 2, the flux prediction module includes a first flux convolution submodule, a first flux downsampling submodule, a second flux convolution submodule, a second flux downsampling submodule, a third flux convolution submodule, a third flux downsampling submodule, a fourth flux convolution submodule, a first flux upsampling submodule, a fifth flux convolution submodule, a second flux upsampling submodule, a sixth flux convolution submodule, a third flux upsampling submodule, a seventh flux convolution submodule, an eighth flux convolution submodule, a fourth flux downsampling submodule, a ninth flux convolution submodule, a fifth flux downsampling submodule, a tenth flux convolution submodule, a sixth flux downsampling submodule, and an eleventh flux convolution submodule, which are arranged in sequence; The projection image is passed through the eighth flux convolution submodule to obtain a thirteenth flux feature map, the thirteenth flux feature map is passed through the fourth flux downsampling submodule to obtain a fourteenth flux feature map, the fourteenth flux feature map is passed through the ninth flux convolution submodule to obtain a fifteenth flux feature map, the fifteenth flux feature map is passed through the fifth flux downsampling submodule to obtain a sixteenth flux feature map, the sixteenth flux feature map is passed through the tenth flux convolution submodule to obtain a seventeenth flux feature map, the seventeenth flux feature map is passed through the sixth flux downsampling submodule to obtain an eighteenth flux feature map, and the eighteenth flux feature map is passed through the eleventh flux convolution submodule to obtain a nineteenth flux feature map; The projection map and the dose map output by the dose prediction module are passed through the first flux convolution submodule to obtain a first flux feature map, the first flux feature map is passed through the first flux downsampling submodule to obtain a second flux feature map, the second flux feature map is passed through the second flux convolution submodule to obtain a third flux feature map, the third flux feature map is passed through the second flux downsampling submodule to obtain a fourth flux feature map, the fourth flux feature map is passed through the third flux convolution submodule to obtain a fifth flux feature map, the fifth flux feature map is passed through the third flux downsampling submodule to obtain a sixth flux feature map, the sixth flux feature map is passed through the fourth flux convolution submodule to obtain a seventh flux feature map, the seventh flux feature map is added to the nineteenth flux feature map and passed through the first flux upsampling submodule to obtain the ninth flux feature map. The eighth flux characteristic map, the eighth flux characteristic map, the fifth flux characteristic map and the eighth dose characteristic map outputted midway by the dose prediction module are spliced ​​and passed through the fifth flux convolution submodule to obtain the ninth flux characteristic map. The ninth flux characteristic map is passed through the second flux upsampling submodule to obtain the tenth flux characteristic map. The tenth flux characteristic map, the third flux characteristic map and the tenth dose characteristic map outputted midway by the dose prediction module are spliced ​​and passed through the sixth flux convolution submodule to obtain the eleventh flux characteristic map. The eleventh flux characteristic map is passed through the third flux upsampling submodule to obtain the twelfth flux characteristic map. The twelfth flux characteristic map, the first flux characteristic map and the twelfth dose characteristic map outputted midway by the dose prediction module are spliced ​​and passed through the seventh flux convolution submodule to obtain a flux map.

6. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 4.

Images