Brain glioma close-range implantation treatment dose prediction method

Abstract

The invention discloses a brain glioma close-range implantation therapeutic dose prediction method, and relates to the technical field of therapeutic dose prediction, and the method comprises the following steps: obtaining patient data and extracting key data of a patient, constructing a network CDC Transform model, training the network CDC Transform model, and outputting a therapeutic dose prediction result through the network CDC Transform model. According to the method, the CDC Transform model based on the Transform model is constructed to realize automatic prediction of the close-range implantation therapeutic dose of the brain glioma, channel space attention is added before the down-sampling process of the original Transform model so as to automatically distribute the weight of the channel and enhance the prediction result, dependence on doctors and physical personal experience is not needed, and the method has the advantages that the method is simple and convenient to operate, and the cost is low. And the dose distribution can be accurately predicted.

Description

Technical Field

[0001] The present invention relates to the technical field of therapeutic dose prediction, and in particular to a method for predicting the therapeutic dose of a brain glioma by close-range implantation. Background Art

[0002] Brachytherapy refers to implanting a radiation source near a tumor in the patient's body, using the radiation source to irradiate the tumor and thus kill cancer cells. Brachytherapy is essentially a precise radiotherapy that uses an implant gun and an implant needle to implant radioactive particles near the tumor, which can effectively kill the tumor and has a lower impact on normal tissue. However, there are also quite a few problems. For example, the preoperative plan and the operation during the operation cannot be fully matched, which is prone to deviations. In addition, the formulation of the surgical plan is closely related to the doctor's clinical experience and subjective factors, and the formulation efficiency is low, and the results may not be the optimal solution. The location and dose calculation of radioactive particles are also the key to the success of particle implantation interventional surgery. Traditional methods make it difficult to make the dose distribution meet clinical requirements, and dose radiation cold and hot areas are prone to occur. Currently, there are few dose prediction methods for brachytherapy.

[0003] In brachytherapy, radiation dose distribution is usually closely related to particle position, number, and activity. The dose distribution decays dramatically as the distance from the particle increases. Therefore, after making a good dose distribution plan, it is also difficult to determine the particle position. At present, there are not many automated prediction methods for dose distribution in brachytherapy. Most of them are still based on the personal experience of doctors and physicists for planning. The calculation to determine the particle position is also very complicated. There is basically no deep learning method for brachytherapy. Therefore, the present invention proposes a dose prediction method for brachytherapy of glioma to solve the problems existing in the prior art. Summary of the invention

[0004] In view of the above problems, the purpose of the present invention is to propose a method for predicting the dose of brachytherapy for glioma, so as to solve the problem that there are not many automated prediction methods for dose distribution of brachytherapy in the prior art, and most of them are still based on the personal experience of doctors and physicists for planning. The calculation to determine the position of particles is also very complicated, and there is basically no deep learning method for brachytherapy.

[0005] In order to achieve the purpose of the present invention, the present invention is implemented by the following technical scheme: a method for predicting the dose of brain glioma brachytherapy, comprising the following steps:

[0006] Step 1: Collect the medical data of glioma patients in the hospital's medical record system, pre-process the collected medical data, and then extract the key data of the patients;

[0007] Step 2: Add channel spatial attention before the downsampling process of the Transformer model to build a network CDC Transformer model based on the Transformer model;

[0008] Step 3: Using the key data of different glioma patients extracted in step 1 as the data set, the network CDC Transformer model constructed in step 2 is trained;

[0009] Step 4: The patient's CT image, dose information, and the brain contour and PTV area drawn by the doctor are used as the input of the network CDC Transformer model. The network CDC Transformer model outputs the patient's 3D dose image to predict the dose of brachytherapy for brain glioma.

[0010] A further improvement is that in step one, the specific steps for preprocessing the medical data are: first, remove erroneous records in the data set by data cleaning, then convert each data into a unified format by data standardization, and then expand the data set by data enhancement technology.

[0011] Further improvement lies in: in the step 1, the specific steps of extracting the patient's key data are: first determine the key fields of the key data, then extract the determined key fields from the preprocessed data set, and then integrate the extracted data into a structured format to obtain key data including CT images, ptv, particle position, dose and external.

[0012] A further improvement is that in step 2, before the downsampling process, the CBAM module is first used to perform channel attention processing on the model to enhance the model's attention to important features, and then the CBAM module is used to perform spatial attention processing on the model.

[0013] A further improvement is that in step 2, during the downsampling process of the Transformer model, a distance attention module is used to perform distance convolution calculation, and a vector aggregation module is added to the Transformer model.

[0014] Further improvements are as follows: in step three, when training the network CDCTransformer model, the model training batch size is set to 2, the initial learning rate is 3e-6, the Adam optimizer is used, the cosine annealing algorithm is used to ensure the training results, and the total number of training iterations is set to 100,000.

[0015] A further improvement is that in step three, the dose of each voxel point is extracted as input during the training of the network CDC Transformer model, so as to enable the network CDC Transformer model to learn the relationship between the dose of each voxel and the particle position.

[0016] A further improvement is that in step 3, the loss function during the training of the network CDCTransformer model is the MSE function, which is expressed by the following formula:

[0017]

[0018] Where i represents each voxel, n represents the total number of voxels, and P i represents the predicted value of each voxel, G i Represents the true dose value of each voxel;

[0019] The model training evaluation indicators use the DICE coefficient and MSE function. The DICE coefficient is expressed as follows:

[0020]

[0021] Among them, X is the actual dose result, and Y is the predicted dose result.

[0022] The beneficial effects of the present invention are as follows: the present invention realizes the automatic prediction of the dose of close-range implantation therapy for brain glioma by constructing a network CDCTransformer model based on the Transformer model, and adds channel spatial attention before the downsampling process of the original Transformer model to automatically allocate channel weights and enhance the prediction results. It does not need to rely on the personal experience of doctors and physicists, and can more accurately predict the dose distribution, help doctors analyze the dose prediction results more intuitively, and assist doctors in formulating radiotherapy plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0024] Figure 1 It is a schematic flow chart of a method for predicting dose of brachytherapy for glioma of the present invention;

[0025] Figure 2 is a visualization diagram of the extracted particle positions in an embodiment of the present invention;

[0026] Figure 3 is a schematic diagram of the structure of a channel space attention module in an embodiment of the present invention;

[0027] Figure 4 is a schematic diagram of the structure of a distance attention module in an embodiment of the present invention;

[0028] Figure 5 is a schematic diagram of the distance calculate calculation process in an embodiment of the present invention;

[0029] Figure 6 is a schematic diagram of the structure of a vector aggregation module in an embodiment of the present invention;

[0030] Figure 7 is a comparison chart of the actual dose and the predicted dose for patient I in an embodiment of the present invention;

[0031] Figure 8 is a comparison chart of the actual dose and the predicted dose for patient II in an embodiment of the present invention;

[0032] Fig. 9 is a DVH map of patient I in an embodiment of the present invention;

[0033] Fig.10 is the DVH diagram of patient II in the embodiment of the present invention. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] See also Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 , Figure 6 , Figure 7 , Figure 8 , Fig. 9 , Fig.10 This embodiment provides a method for predicting the dose of a glioma brachytherapy, comprising the following steps:

[0036] Step 1: Obtain patient data and extract key patient data

[0037] The medical data of fifty patients with brain glioma were collected from the hospital's medical record system, and the collected medical data were preprocessed. The specific steps were as follows: first, data cleaning was used to remove erroneous records in the data set, and then data standardization was used to convert the data into a unified format for subsequent analysis. Then, data enhancement technology was used to expand the data set to improve the generalization ability of the subsequent model.

[0038] Then extract the key data of the patient from the preprocessed data set. The specific steps are: first determine the key fields of the key data, then extract the determined key fields from the preprocessed data set, and then integrate the extracted data into a structured format, such as a CSV file or a database table, to obtain key data including CT images, ptv (planned target volume), particle position, dose (dose distribution), and external (external information);

[0039] In view of the high influence of the number and position of particles on the dose distribution, this embodiment extracts the particle position and number information from the patient's medical data as a channel input. The extracted particle position visualization diagram is shown in FIG. Figure 2 ;

[0040] Step 2: Build a network CDC Transformer model

[0041] Before the downsampling process of the Transformer model, channel spatial attention is added to build a network CDC (Change Detection and Classification) Transformer model based on the Transformer model to automatically assign channel weights and enhance the prediction results. The channel spatial attention module consists of a channel attention module and a spatial attention module. For its structure diagram, see Figure 3 ,The channel attention module of the channel spatial attention module can assign weights by itself, and can assign larger training weights to important channels. In this way, the model can determine important input channels by itself and adjust weights to improve training effects. The spatial attention module can better capture spatial information. Due to the complex structure of the brain, improving the ability to capture spatial information can make the model learn better;

[0042] In this embodiment, before the downsampling process, the CBAM module (convolutional block attention module) is first used to perform channel attention processing on the model to enhance the model's attention to important features, and then the CBAM module is used to perform spatial attention processing on the model, which helps to improve the performance of the model;

[0043] In the downsampling process of the Transformer model, this embodiment uses a distance attention module to perform distance convolution calculation (i.e., multiplying each value on the original voxel by an inverse square coefficient to transform the relationship between each voxel and the particle voxel distance), and adds a vector aggregation module to the model to integrate different features into a feature vector for characterizing the probability of the existence of the feature and the spatial structure. The distance attention module integrates a formula about the inverse distance to assist the particle position input. The module only integrates an inverse relationship, and the remaining parameters are learned by the model itself during training, so that the model can accurately learn the distance-related dose distribution information;

[0044] The distance attention module structure can be found in Figure 4 , where the distance calculate process is shown in Figure 5 , that is, since the dose distribution of particle radiation is inversely proportional to the distance between particles, this embodiment embeds an inverse relationship formula, and the obtained output feature is convolved with the original distance_map to obtain the distance feature, and then combined with the features of the previous part of the model to become a new feature, which is input to the next layer of the network, that is, the initial dose is a dose value at each voxel point, which is converted into a new voxel value through the inverse formula calculation, and then other steps are performed. In this way, the model can learn the relationship between the particle position and the distance between each position;

[0045] Vector aggregation module structure diagram see Figure 6 The vector aggregation module can list the feature values ​​of each direction of the bottom-level feature map as vector calculations. The modulus of the vector represents the probability of the feature's existence, and the angle of the vector represents the spatial structure of the feature. This allows the model to learn the spatial features more clearly, and the model can not only predict the dose distribution, but also the position of the particles, which can reduce the cost of the doctor's particle position calculation;

[0046] Step 3: Train the network CDCTransformer model

[0047] The key data of fifty patients with glioma extracted in step 1 were used as the data set to train the network CDCTransformer model constructed in step 2. During the training process, the dose of each voxel point was extracted as input for the network CDCTransformer model to learn the relationship between the dose of each voxel and the particle position. When training the prediction model, the parameters were set as follows: the model training batch size was set to 2, the initial learning rate was 3e-6, the Adam optimizer was used, the cosine annealing algorithm was used to ensure the training results, the total number of training iterations was set to 100,000, and the NVIDIA GeForce3080TI GPU and PyTorch1.70 framework were used for training;

[0048] In this embodiment, the loss function in the network CDCTransformer model training process is the MSE function, and the MSE function is expressed by the following formula:

[0049]

[0050] Where i represents each voxel, n represents the total number of voxels, and P i represents the predicted value of each voxel, G i Represents the true dose value of each voxel;

[0051] The model training evaluation indicators use the DICE coefficient and MSE function. The DICE coefficient is expressed as follows:

[0052]

[0053] Among them, X is the actual dose result, and Y is the predicted dose result

[0054] Step 4: Output the treatment dose prediction results through the network CDCTransformer model

[0055] The input of the network CDC Transformer model is three channels. In this embodiment, the patient's CT image, dose information, and the cranial contour and PTV area drawn by the doctor are used as the input of the network CDC Transformer model. The network CDC Transformer model outputs the patient's 3D dose image to predict the dose of close-range implantation therapy for brain glioma. Then, the patient's 3D dose image is superimposed on the original CT image to obtain a dose distribution map, which helps doctors analyze the dose prediction results more intuitively.

[0056] The output of the network CDC Transformer model in this embodiment is a three-dimensional dose distribution. Two patients (patient I and patient II) are randomly selected and the visualized dose distribution results are compared. The actual dose and predicted dose comparison of patient I are shown in Figure 7 ( Figure 7 The actual dose is on the left and the predicted dose is on the right). For the comparison chart of the actual dose and predicted dose of patient II, see Figure 8 ( Figure 8 The left side is the actual dose, and the right side is the predicted dose). The DVH diagram of patient I can be found in Fig. 9 , the DVH diagram of patient II can be found in Fig.10 , where the solid line is the actual DVH curve of the patient, and the dotted line is the predicted DVH curve of the patient. It can be seen from the figure that the prediction error is very small.

[0057] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting dose of brachytherapy for glioma, characterized in that: The following steps are involved: Step 1: Collect the medical data of glioma patients in the hospital's medical record system, pre-process the collected medical data, and then extract the key data of the patients; Step 2: Add channel spatial attention before the downsampling process of the Transformer model to build a network CDC Transformer model based on the Transformer model; Step 3: Using the key data of different glioma patients extracted in step 1 as the data set, the network CDC Transformer model constructed in step 2 is trained; Step 4: The patient's CT image, dose information, and the brain contour and PTV area drawn by the doctor are used as the input of the network CDC Transformer model. The network CDC Transformer model outputs the patient's 3D dose image to predict the dose of brachytherapy for brain glioma.

2. A method for predicting dose of brachytherapy for glioma according to claim 1, characterized in that: In the step 1, the specific steps of preprocessing the medical data are: first, remove the erroneous records in the data set by using data cleaning, then convert each data into a unified format by using data standardization, and then expand the data set by using data enhancement technology.

3. The method for predicting dose of brachytherapy for glioma according to claim 1, characterized in that: In the step 1, the specific steps of extracting the patient's key data are: first determine the key fields of the key data, then extract the determined key fields from the preprocessed data set, and then integrate the extracted data into a structured format to obtain key data including CT images, ptv, particle position, dose and external.

4. A method for predicting dose of brachytherapy for glioma according to claim 1, characterized in that: In the step 2, before the downsampling process, the CBAM module is first used to perform channel attention processing on the model to enhance the model's attention to important features, and then the CBAM module is used to perform spatial attention processing on the model.

5. The method for predicting dose of brachytherapy for glioma according to claim 1, characterized in that: In the step 2, during the downsampling process of the Transformer model, a distance attention module is used to perform distance convolution calculation, and a vector aggregation module is added to the Transformer model.

6. A method for predicting dose of brachytherapy for glioma according to claim 1, characterized in that: In step 3, when training the network CDC Transformer model, the model training batch size is set to 2, the initial learning rate is set to 3e-6, the Adam optimizer is used, the cosine annealing algorithm is used to ensure the training results, and the total number of training iterations is set to 100,000.

7. A method for predicting dose of brachytherapy for glioma according to claim 1, characterized in that: In the step three, the dose of each voxel point is extracted as input during the training process of the network CDC Transformer model, and is used for the network CDC Transformer model to learn the relationship between the dose of each voxel and the particle position.

8. The method for predicting dose of brachytherapy for glioma according to claim 1, characterized in that: In the step 3, the loss function in the network CDC Transformer model training process is the MSE function, and the MSE function is expressed by the following formula: Where i represents each voxel, n represents the total number of voxels, and P i represents the predicted value of each voxel, G i Represents the true dose value of each voxel; The model training evaluation indicators use the DICE coefficient and MSE function. The DICE coefficient is expressed as follows: Among them, X is the actual dose result, and Y is the predicted dose result.

Images