CN117612726B - Method, device, equipment and medium for predicting radiation pneumonitis - Google Patents
Method, device, equipment and medium for predicting radiation pneumonitis Download PDFInfo
- Publication number
- CN117612726B CN117612726B CN202311281348.8A CN202311281348A CN117612726B CN 117612726 B CN117612726 B CN 117612726B CN 202311281348 A CN202311281348 A CN 202311281348A CN 117612726 B CN117612726 B CN 117612726B
- Authority
- CN
- China
- Prior art keywords
- features
- prediction
- dosimetry
- dose volume
- volume histogram
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 206010037765 Radiation pneumonitis Diseases 0.000 title claims abstract description 59
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000004980 dosimetry Methods 0.000 claims abstract description 71
- 206010035664 Pneumonia Diseases 0.000 claims abstract description 36
- 238000013136 deep learning model Methods 0.000 claims abstract description 18
- 238000000605 extraction Methods 0.000 claims description 19
- 206010028980 Neoplasm Diseases 0.000 claims description 17
- 230000002776 aggregation Effects 0.000 claims description 15
- 238000004220 aggregation Methods 0.000 claims description 15
- 238000013528 artificial neural network Methods 0.000 claims description 13
- 230000004044 response Effects 0.000 claims description 13
- 238000007408 cone-beam computed tomography Methods 0.000 claims description 11
- 238000013527 convolutional neural network Methods 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 9
- 230000004931 aggregating effect Effects 0.000 claims description 5
- 206010058467 Lung neoplasm malignant Diseases 0.000 description 14
- 201000005202 lung cancer Diseases 0.000 description 14
- 208000020816 lung neoplasm Diseases 0.000 description 14
- 238000001959 radiotherapy Methods 0.000 description 13
- 239000013598 vector Substances 0.000 description 12
- 238000012545 processing Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 5
- 230000004927 fusion Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 230000004913 activation Effects 0.000 description 3
- 238000011176 pooling Methods 0.000 description 3
- 201000011510 cancer Diseases 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 210000004072 lung Anatomy 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 206010061218 Inflammation Diseases 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000004054 inflammatory process Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000006116 polymerization reaction Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/80—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for detecting, monitoring or modelling epidemics or pandemics, e.g. flu
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- Theoretical Computer Science (AREA)
- Primary Health Care (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Radiation-Therapy Devices (AREA)
Abstract
The application relates to the technical field of medical treatment, in particular to a method, a device, equipment and a medium for predicting radiation pneumonitis. The method comprises the following steps: acquiring a dose volume histogram and dosimetry information; inputting the dose volume histogram and the dosimetry information into a preset pneumonia prediction model to obtain a prediction result; wherein the pneumonia prediction model is a multi-modal deep learning model with a hierarchical network. With this arrangement, the prediction result can be obtained in advance, and then the pretreatment or early treatment of radiation pneumonitis can be performed based on the prediction result.
Description
Technical Field
The application relates to the technical field of medical treatment, in particular to a method, a device, equipment and a medium for predicting radiation pneumonitis.
Background
Lung cancer is one of the malignant tumors with highest incidence in China, and radiotherapy is an important treatment method. Various degrees of radiation pneumonitis can occur after radiation treatment of lung cancer.
In the existing treatment process, the radiation pneumonitis needs to be treated when the radiation pneumonitis appears after the lung cancer treatment of the user is finished, but in the mode, the radiation pneumonitis cannot be intervened in advance, so that the treatment effect is poor.
Disclosure of Invention
In view of the foregoing, embodiments of the present application are directed to a radiation pneumonitis prediction method, apparatus, device, and medium for predicting radiation pneumonitis in advance, so as to facilitate advanced treatment.
The first aspect of the application provides a method for predicting radiation pneumonitis, comprising the following steps:
Acquiring a dose volume histogram and dosimetry information;
inputting the dose volume histogram and the dosimetry information into a preset pneumonia prediction model to obtain a prediction result;
Wherein the pneumonia prediction model is a multi-modal deep learning model with a hierarchical network.
In some embodiments, the pneumonia prediction model comprises: the device comprises a feature extraction module, a feature aggregation module, a feature weight module and a prediction module;
the feature extraction module is used for extracting features of the dose volume histogram to obtain corresponding features of the dose volume histogram, extracting features of the dosimetry information and obtaining the dosimetry features;
the feature aggregation module is used for aggregating the dosimetry features and the dose volume histogram features to obtain multi-modal features;
The feature severity module is used for fusing the polymerized dosimetry features and the dose volume histogram features in the multi-modal features to obtain fused multi-modal features;
the prediction module is used for predicting based on the fused multi-mode characteristics to obtain a prediction result;
in some embodiments, the feature extraction module comprises: a 3D convolutional neural network and a fully-connected neural network;
The 3D convolutional neural network is used for extracting the characteristics of the dose volume histogram to obtain corresponding dose volume histogram characteristics;
The fully-connected neural network is used for extracting characteristics of the dosimetry information to obtain the dosimetry characteristics.
In some embodiments, the prediction result comprises: grade of radiation pneumonitis;
the prediction module is used for predicting based on the fused multi-mode characteristics to obtain the level of the radiation pneumonitis.
In some embodiments, a method of pre-training a pneumonia prediction model comprises:
acquiring information of a preset number of cases of radiation pneumonitis as sample information
Training through the sample information device by a deep learning model built in advance to obtain a pneumonia prediction model.
In some embodiments, further comprising:
Acquiring a CBCT image, an MVCT image and dose information;
And inputting the CBCT image, the MVCT image and the dose information into a preset tumor response prediction model to obtain a tumor response prediction result.
In some embodiments, the tumor response prediction model is a transducer model.
A second aspect of the present application provides a radiation pneumonitis prediction device comprising:
The acquisition module is used for acquiring a dose volume histogram and dosimetry information;
The prediction module is used for inputting the dose volume histogram and the dosimetry information into a preset pneumonia prediction model to obtain a prediction result;
Wherein the pneumonia prediction model is a multi-modal deep learning model with a hierarchical network.
A third aspect of the present application provides an electronic apparatus, comprising:
A processor and a memory for storing a program executable by the processor;
the processor is used for realizing the radiation pneumonitis prediction method by running the program in the memory.
A fourth aspect of the application provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, causes the processor to perform the above-described radiation pneumonitis prediction method.
The application provides a method for predicting radiation pneumonitis, which is used for acquiring a dose volume histogram and dosimetry information; inputting the dose volume histogram and the dosimetry information into a preset pneumonia prediction model to obtain a prediction result; wherein the pneumonia prediction model is a multi-modal deep learning model with a hierarchical network. With this arrangement, the prediction result can be obtained in advance, and then the pretreatment or early treatment of radiation pneumonitis can be performed based on the prediction result.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing embodiments of the present application in more detail with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and together with the embodiments of the application, and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.
Fig. 1 is a flow chart of a method for predicting radiation pneumonitis according to an embodiment of the present application.
Fig. 2 is a schematic diagram of a pneumonia prediction model according to an embodiment of the present application.
Fig. 3 is a schematic structural view of a radiation pneumonitis prediction device according to an embodiment of the present application.
Fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Summary of the application
Lung cancer is one of the malignant tumors with highest incidence in China, and radiotherapy is an important treatment method. Various degrees of radiation pneumonitis can occur after radiation treatment of lung cancer. In the existing treatment process, the radiation pneumonitis needs to be treated when the radiation pneumonitis appears after the lung cancer treatment of the user is finished, but in the mode, the radiation pneumonitis cannot be intervened in advance, so that the treatment effect is poor.
In order to solve the above problems, the present application provides a solution to acquire a dose volume histogram and dosimetry information; inputting the dose volume histogram and the dosimetry information into a preset pneumonia prediction model to obtain a prediction result; wherein the pneumonia prediction model is a multi-modal deep learning model with a hierarchical network. With this arrangement, the prediction result can be obtained in advance, and then the pretreatment or early treatment of radiation pneumonitis can be performed based on the prediction result.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Exemplary method
Fig. 1 is a flow chart of a method for predicting radiation pneumonitis according to an embodiment of the present application. As shown in fig. 1, the method includes the following.
Step S110, acquiring a dose volume histogram and dosimetry information;
The dose volume histogram and the dosimetry information refer to the dose volume histogram and the dosimetry information during lung cancer treatment; the metering information comprises dosage information, a dosage dividing mode and a divided dosage size in the lung cancer process.
The specific way to obtain the dose volume histogram and the dosimetry information may be to obtain the dose volume histogram and the dosimetry information recorded during the lung cancer treatment. In practice, relevant data (including dose volume histogram and dosimetry information) is recorded by relevant units or individuals such as hospitals or doctors during the course of treating lung cancer in patients. Thus, the method can be directly obtained in relevant units or individuals such as hospitals or doctors when the dose volume histogram and the dosimetry information need to be obtained.
Step S120, inputting the dose volume histogram and the dosimetry information into a preset pneumonia prediction model to obtain a prediction result;
Wherein the pneumonia prediction model is a multi-modal deep learning model with a hierarchical network.
With this arrangement, the prediction result can be obtained in advance, and then the pretreatment or early treatment of radiation pneumonitis can be performed based on the prediction result.
With further reference to fig. 2, the pneumonia prediction model includes: the device comprises a feature extraction module, a feature aggregation module, a feature weight module and a prediction module;
the feature extraction module is used for extracting features of the dose volume histogram to obtain corresponding features of the dose volume histogram, extracting features of the dosimetry information and obtaining the dosimetry features; specifically, the feature extraction module includes: uppermost 1DMLP and 3DCNN in fig. 2.
The feature aggregation module is used for aggregating the dosimetry features and the dose volume histogram features to obtain multi-modal features; specifically, the feature extraction module includes: the first two transducers (i.e., the two transducer layers) from top to bottom in fig. 2.
The feature severity module is used for fusing the polymerized dosimetry features and the dose volume histogram features in the multi-modal features to obtain fused multi-modal features; specifically, the feature extraction module includes: the lowermost transducer in fig. 2 and the softmax layer next to this transducer.
The prediction module is used for predicting based on the fused multi-mode characteristics to obtain a prediction result. Specifically, the feature extraction module includes: the lowest softmax layer in fig. 2.
Thus, fusing the multi-modal features (dose volume histogram features and dosimetry features) through the various modules described above allows for more accurate predicted structures.
Specifically, the feature extraction module includes: a 3D convolutional neural network and a fully-connected neural network;
The 3D convolutional neural network is used for extracting the characteristics of the dose volume histogram to obtain corresponding dose volume histogram characteristics;
The fully-connected neural network is used for extracting characteristics of the dosimetry information to obtain the dosimetry characteristics.
Specifically, a Dose Volume Histogram (DVH) and a dosimetry characteristic are acquired for a lung cancer patient adopting a conventional dose segmentation mode (56-60 Gy/27-30 times) or stereotactic radiotherapy (42-50 Gy/6-10 times), and are input into a multi-modal grading network. Since the DVH and the dosimetry features belong to different modalities, two independent processing branches are adopted for feature learning in order to learn the targeted DVH features and the dosimetry features respectively. And for DVH processing branches, adopting a plurality of convolution layers and downsampling layers which are continuously stacked by a 3D convolution neural network to learn DVH characteristics layer by layer. During this process, the DVH feature remains in three-dimensional form. For the dosimetry feature processing branch, a one-dimensional fully connected neural network (multi-layer perceptron, MLP) is adopted to perform the dosimetry feature learning layer by layer. During this process, the dosimetry features remain in one-dimensional form.
In order to accurately predict the patient from the radiation pneumonitis, the three-dimensional DVH features and the one-dimensional dosimetry features extracted by the two branches are required to be effectively fused, and the features of the two different modes are required to be dimension matched. For this purpose, firstly, three-dimensional DVH features are subjected to a spatial global averaging pooling operation, and feature vectors fdvh (dose volume histogram features) generated after pooling are all 1×d in scale. The dosimetry information is processed by a one-dimensional fully connected neural network to obtain a dosimetry characteristic, and the dosimetry characteristic is finally converted into a characteristic vector f dos (dosimetry characteristic) of 1x d.
Then, feature aggregation and fusion are performed on the 1×d feature vectors f dvh and f dos generated by the two modes. The operation flow of multi-mode feature aggregation and fusion is shown in fig. 2, and the structure of the multi-mode feature aggregation and fusion comprises two parts, wherein one part is a feature aggregation module, and the other part is a feature weight measurement module.
The polymerization process is as follows: the two paths of 1×d eigenvectors generated by the DVH and the dosimetry information respectively pass through two linear connection layers and keep the vector size unchanged, and the eigenvalue is still 1×d. And then, carrying out layer normalization processing on the generated two paths of feature vectors so as to accelerate the training of the model and avoid the overfitting of the model. And stacking the two paths of feature vectors subjected to layer normalization, and performing dimension transformation on the feature vectors to 2 x d to obtain the multi-modal feature.
And then carrying out feature weight measurement (i.e. carrying out feature fusion), learning the weight when the dose volume histogram features and the dosimetry features are fused by utilizing the aggregated multi-modal features, and representing the importance degree of the features of two different modes to the final classification of the radiation pneumonitis, so as to carry out weight measurement on the features of the two different modes, and then converting the multi-modal features with the dimension of 2x d into the multi-modal features with the dimension of 1 x d based on the weight obtained after the weight measurement.
Specifically, the feature vector after aggregation is subjected to continuous processing of a linear connection layer, a tanh activation function and a linear connection layer, and features are converted into weight vectors with the sum of 1 and the dimension of 2×1 through a softmax layer, so that the weight corresponding to the dose volume histogram features and the dosimetry features is represented.
Further, the prediction result includes: grade of radiation pneumonitis;
the prediction module is used for predicting based on the fused multi-mode characteristics to obtain the level of the radiation pneumonitis.
It should be noted that, in order to better characterize the actual condition of pneumonia, the pneumonia may be preferentially classified. Namely: different pneumonia cases correspond to different levels of radiation pneumonitis.
In this way, the prediction module may input the generated two paths of 1×d feature vectors into two linear connection layers respectively, to obtain feature vectors with dimensions of m, where m is the number of classes to be classified. The generated m-dimensional features with the size of two paths are subjected to further feature fusion through global average pooling, and then the classification of the radiation pneumonitis can be realized through Sigmod activation function operation. Here, for each level of radiation pneumonitis, the probability belonging to that level is determined by Sigmod activation functions based on the weighted feature vectors described above.
Further, the application also provides a method for pre-training a pneumonia prediction model, which comprises the following steps:
acquiring information of a preset number of cases of radiation pneumonitis as sample information
Training through the sample information device by a deep learning model built in advance to obtain a pneumonia prediction model.
Specifically, in the scheme provided by the application, the multi-mode hierarchical network is used for building the deep learning model. According to the pre-collected data and the preliminary test result, relevant parameters are adjusted and optimized, and training of a deep learning model is carried out to obtain a pneumonia prediction model so as to establish the deep learning model for predicting the risk of the Radiation Pneumonitis (RP) in the radiation therapy of the lung cancer in different segmentation modes by using a Dose Volume Histogram (DVH) and a dosimetry characteristic.
In some embodiments, the present application provides a method comprising:
Acquiring a CBCT image, an MVCT image and dose information;
And inputting the CBCT image, the MVCT image and the dose information into a preset tumor response prediction model to obtain a tumor response prediction result.
Wherein the tumor response prediction model is a transducer model. Specifically, the study is to develop early tumor response for lung cancer patients undergoing radiotherapy (56-60 Gy/27-30 times) by adopting a conventional dose division mode, extract tumor information of the patients by using CBCT images or MVCT images (respectively 5 times, 10 times, 15 times, 20 times and 25 times) during radiotherapy of the patients and CT images of the patients within 6 months after the radiotherapy is finished, and establish and optimize a change model based on CBCT images or MVCT images during the 5 th and 10 th radiotherapy for 15 times, 20 times, 25 times and 3 months and 6 months after the radiotherapy is finished by using deep learning (Tranformer network). The accuracy of the model was verified by using parameters of the test set tumor volume, mean distance between tumor volumes (DTA) and tumor structure uniformity coefficient (SSIM) and the like.
Furthermore, the application can also be used for carrying out radiation pneumonitis prediction research on lung cancer patients adopting stereotactic radiotherapy (42-50 Gy/6-10 times). Radiation therapy planning is extracted by Python software and includes Mean Lung Dose (MLD), generalized equivalent uniform dose (gEUD), DVH features of the relative volumes of the lungs that receive more than a certain dose, and dose distribution features containing first order statistics and texture features. In combination with clinical data such as CBCT images or MVCT images during radiotherapy of a patient, CT images of a patient within 6 months after the end of radiotherapy, and an index of examined inflammation, a deep learning model (multi-modal hierarchical network) is constructed that predicts the occurrence of radiation pneumonitis of different degrees based on a DVH having a DVH, a dose profile corrected to EQD2, a dosimetry, and a dose profile corrected to EQD 2.
In the scheme provided by the application, the related content of the radiation pneumonitis prediction method and the tumor response prediction is provided based on the thought of 'predicting through a deep learning model'.
Exemplary apparatus
The device embodiment of the application can be used for executing the method embodiment of the application. For details not disclosed in the embodiments of the apparatus of the present application, please refer to the embodiments of the method of the present application.
Fig. 3 is a block diagram of a radiation pneumonitis prediction device according to an embodiment of the present application. As shown in fig. 3, the apparatus includes:
An acquisition module 31 for acquiring a dose volume histogram and dosimetry information;
the prediction module 32 is configured to input the dose volume histogram and the dosimetry information into a preset pneumonia prediction model to obtain a prediction result;
Wherein the pneumonia prediction model is a multi-modal deep learning model with a hierarchical network.
In some embodiments, the pneumonia prediction model comprises: the device comprises a feature extraction module, a feature aggregation module, a feature weight module and a prediction module;
the feature extraction module is used for extracting features of the dose volume histogram to obtain corresponding features of the dose volume histogram, extracting features of the dosimetry information and obtaining the dosimetry features;
the feature aggregation module is used for aggregating the dosimetry features and the dose volume histogram features to obtain multi-modal features;
The feature severity module is used for fusing the polymerized dosimetry features and the dose volume histogram features in the multi-modal features to obtain fused multi-modal features;
the prediction module is used for predicting based on the fused multi-mode characteristics to obtain a prediction result;
in some embodiments, the feature extraction module comprises: a 3D convolutional neural network and a fully-connected neural network;
The 3D convolutional neural network is used for extracting the characteristics of the dose volume histogram to obtain corresponding dose volume histogram characteristics;
The fully-connected neural network is used for extracting characteristics of the dosimetry information to obtain the dosimetry characteristics.
In some embodiments, the prediction result comprises: grade of radiation pneumonitis;
the prediction module is used for predicting based on the fused multi-mode characteristics to obtain the level of the radiation pneumonitis.
In some embodiments, a method of pre-training a pneumonia prediction model comprises:
acquiring information of a preset number of cases of radiation pneumonitis as sample information
Training through the sample information device by a deep learning model built in advance to obtain a pneumonia prediction model.
Exemplary electronic device
Next, an electronic device according to an embodiment of the present application is described with reference to fig. 4. Fig. 4 illustrates a block diagram of an electronic device according to an embodiment of the application.
As shown in fig. 4, electronic device 400 includes one or more processors 410 and memory 420.
Processor 410 may be a Central Processing Unit (CPU) or other form of processing unit having data processing and/or instruction execution capabilities and may control other components in electronic device 400 to perform desired functions.
Memory 420 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, random Access Memory (RAM) and/or cache memory (cache), and the like. The non-volatile memory may include, for example, read Only Memory (ROM), hard disk, flash memory, and the like. One or more computer program instructions may be stored on the computer readable storage medium that can be executed by the processor 410 to implement the radiation pneumonitis prediction method and/or other desired functions of the various embodiments of the present application described above. Various contents such as category correspondence may also be stored in the computer-readable storage medium.
In one example, the electronic device 400 may further include: input device 430 and output device 440, which are interconnected by a bus system and/or other form of connection mechanism (not shown).
In addition, the input device 430 may also include, for example, a keyboard, mouse, interface, etc. The output device 440 may output various information including analysis results and the like to the outside. The output device 440 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, etc.
Of course, only some of the components of the electronic device that are relevant to the present application are shown in fig. 4 for simplicity, components such as buses, input/output interfaces, etc. being omitted. In addition, the electronic device may include any other suitable components depending on the particular application.
Exemplary computer program product and computer readable storage Medium
In addition to the methods and apparatus described above, embodiments of the application may also be a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform the steps in a method of predicting radiation pneumonitis according to various embodiments of the application described in the "exemplary methods" section of this specification.
The computer program product may write program code for performing operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the present application may also be a computer-readable storage medium, having stored thereon computer program instructions, which when executed by a processor, cause the processor to perform the steps in a radiation pneumonitis prediction method according to various embodiments of the present application described in the above "exemplary methods" section of the present specification.
The computer readable storage medium may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
Claims (7)
1. A method for predicting radiation pneumonitis, comprising:
Acquiring a dose volume histogram and dosimetry information;
inputting the dose volume histogram and the dosimetry information into a preset pneumonia prediction model to obtain a prediction result;
wherein the pneumonia prediction model is a multi-modal deep learning model with a hierarchical network;
The pneumonia prediction model includes: the device comprises a feature extraction module, a feature aggregation module, a feature weight module and a prediction module;
the feature extraction module is used for extracting features of the dose volume histogram to obtain corresponding features of the dose volume histogram, extracting features of the dosimetry information and obtaining the dosimetry features;
the feature aggregation module is used for aggregating the dosimetry features and the dose volume histogram features to obtain multi-modal features;
The feature severity module is used for fusing the polymerized dosimetry features and the dose volume histogram features in the multi-modal features to obtain fused multi-modal features;
the prediction module is used for predicting based on the fused multi-mode characteristics to obtain a prediction result;
the feature extraction module includes: a 3D convolutional neural network and a fully-connected neural network;
The 3D convolutional neural network is used for extracting the characteristics of the dose volume histogram to obtain corresponding dose volume histogram characteristics;
the fully-connected neural network is used for extracting characteristics of the dosimetry information to obtain the dosimetry characteristics;
The radiation pneumonitis prediction method further comprises the following steps:
Acquiring a CBCT image, an MVCT image and dose information;
And inputting the CBCT image, the MVCT image and the dose information into a preset tumor response prediction model to obtain a tumor response prediction result.
2. The method for predicting radiation pneumonitis according to claim 1, wherein the prediction result comprises: grade of radiation pneumonitis;
the prediction module is used for predicting based on the fused multi-mode characteristics to obtain the level of the radiation pneumonitis.
3. The method of claim 1, wherein the method of pre-training the pneumonia prediction model comprises:
acquiring information of a preset number of cases of radiation pneumonitis as sample information
Training through the sample information device by a deep learning model built in advance to obtain a pneumonia prediction model.
4. The method of claim 1, wherein the tumor response prediction model is a transducer model.
5. A radiation pneumonitis prediction device, comprising:
The acquisition module is used for acquiring a dose volume histogram and dosimetry information;
The prediction module is used for inputting the dose volume histogram and the dosimetry information into a preset pneumonia prediction model to obtain a prediction result;
wherein the pneumonia prediction model is a multi-modal deep learning model with a hierarchical network;
The pneumonia prediction model includes: the device comprises a feature extraction module, a feature aggregation module, a feature weight module and a prediction module;
the feature extraction module is used for extracting features of the dose volume histogram to obtain corresponding features of the dose volume histogram, extracting features of the dosimetry information and obtaining the dosimetry features;
the feature aggregation module is used for aggregating the dosimetry features and the dose volume histogram features to obtain multi-modal features;
The feature severity module is used for fusing the polymerized dosimetry features and the dose volume histogram features in the multi-modal features to obtain fused multi-modal features;
the prediction module is used for predicting based on the fused multi-mode characteristics to obtain a prediction result;
the feature extraction module includes: a 3D convolutional neural network and a fully-connected neural network;
The 3D convolutional neural network is used for extracting the characteristics of the dose volume histogram to obtain corresponding dose volume histogram characteristics;
the fully-connected neural network is used for extracting characteristics of the dosimetry information to obtain the dosimetry characteristics;
The radiation pneumonitis prediction method further comprises the following steps:
Acquiring a CBCT image, an MVCT image and dose information;
And inputting the CBCT image, the MVCT image and the dose information into a preset tumor response prediction model to obtain a tumor response prediction result.
6. An electronic device, comprising:
A processor and a memory for storing a program executable by the processor;
the processor is configured to implement the radiation pneumonitis prediction method according to any one of claims 1 to 4 by running a program in the memory.
7. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program which, when executed by a processor, causes the processor to perform the radiation pneumonitis prediction method of any of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311281348.8A CN117612726B (en) | 2023-09-21 | 2023-09-21 | Method, device, equipment and medium for predicting radiation pneumonitis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311281348.8A CN117612726B (en) | 2023-09-21 | 2023-09-21 | Method, device, equipment and medium for predicting radiation pneumonitis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117612726A CN117612726A (en) | 2024-02-27 |
| CN117612726B true CN117612726B (en) | 2024-09-13 |
Family
ID=89943118
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311281348.8A Active CN117612726B (en) | 2023-09-21 | 2023-09-21 | Method, device, equipment and medium for predicting radiation pneumonitis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117612726B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108766563A (en) * | 2018-05-25 | 2018-11-06 | 戴建荣 | Radiotherapy prediction of result method and system based on dosage group |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015176011A1 (en) * | 2014-05-15 | 2015-11-19 | The Johns Hopkins University | Method, system and computer-readable media for treatment plan risk analysis |
| KR102020383B1 (en) * | 2016-11-28 | 2019-09-10 | 고려대학교산학협력단 | Evaluation method for radiotherapy plan using data-mining |
| US20220384039A1 (en) * | 2021-05-27 | 2022-12-01 | Gsi Technology Inc. | Cancer type prediction system |
| CN114496239B (en) * | 2021-12-03 | 2024-05-14 | 汕头大学医学院附属肿瘤医院 | Novel lung cancer SBRT (styrene butadiene styrene) radiation pneumonitis risk prediction factor and construction method thereof |
| CN114155934B (en) * | 2021-12-10 | 2024-05-14 | 中南大学 | Tumor intensity modulated radiotherapy dosage prediction method based on deep learning |
| WO2023108526A1 (en) * | 2021-12-16 | 2023-06-22 | 中国科学院深圳先进技术研究院 | Medical image segmentation method and system, and terminal and storage medium |
-
2023
- 2023-09-21 CN CN202311281348.8A patent/CN117612726B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108766563A (en) * | 2018-05-25 | 2018-11-06 | 戴建荣 | Radiotherapy prediction of result method and system based on dosage group |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117612726A (en) | 2024-02-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11182894B2 (en) | Method and means of CAD system personalization to reduce intraoperator and interoperator variation | |
| US10867385B2 (en) | Systems and methods for segmentation of intra-patient medical images | |
| US10765888B2 (en) | System and method for automatic treatment planning | |
| CN109069858B (en) | Radiotherapy system and computer readable storage device | |
| CN108766563A (en) | Radiotherapy prediction of result method and system based on dosage group | |
| CN112614133B (en) | Three-dimensional pulmonary nodule detection model training method and device without anchor point frame | |
| An et al. | Medical Image Classification Algorithm Based on Visual Attention Mechanism‐MCNN | |
| US20230112591A1 (en) | Machine learning based medical data checker | |
| Agnello et al. | Unsupervised tissue classification of brain MR images for voxel‐based morphometry analysis | |
| CN116864139A (en) | Disease risk assessment method, device, computer equipment and readable storage medium | |
| US20210304896A1 (en) | Systems and methods for medical diagnosis | |
| Tian et al. | Radiomics and its clinical application: artificial intelligence and medical big data | |
| Wang et al. | Multi-view fusion segmentation for brain glioma on CT images | |
| Jain et al. | Early detection of brain tumor and survival prediction using deep learning and an ensemble learning from radiomics images | |
| Ma et al. | Autism Spectrum Disorder Classification in Children based on Structural MRI Features Extracted using Contrastive Variational Autoencoder | |
| CN117612726B (en) | Method, device, equipment and medium for predicting radiation pneumonitis | |
| Zeng et al. | TransQA: deep hybrid transformer network for measurement-guided volumetric dose prediction of pre-treatment patient-specific quality assurance | |
| Vazquez et al. | A deep learning-based approach for statistical robustness evaluation in proton therapy treatment planning: a feasibility study | |
| CN114462522A (en) | Lung cancer life cycle prediction model training and prediction method, system, device and medium | |
| CN114036755A (en) | Effect prediction method and device, storage medium and electronic equipment | |
| Ummah et al. | Covid-19 and Tuberculosis Detection in X-Ray of Lung Images with Deep Convolutional Neural Network. | |
| CN114121213A (en) | Anesthesia medicine information rechecking method and device, electronic equipment and storage medium | |
| Priya et al. | An intellectual caries segmentation and classification using modified optimization-assisted transformer denseUnet++ and ViT-based multiscale residual denseNet with GRU | |
| Al-Jammas et al. | Brain tumors recognition based on deep learning | |
| Hossain et al. | A hybrid neural network model for brain tumor detection in brain MRI images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |