A data processing system and method for radiotherapy

Abstract

The present invention relates to the field of medical data processing technology, and specifically to a data processing system and method for radiotherapy, including a data acquisition and processing module, a data synchronization mapping module, a data cleaning and annotation module, a data analysis and optimization module, and a data visualization module; wherein: the data acquisition and processing module is used to acquire multi-source data from different devices; the data synchronization mapping module is used to synchronize and align data from different sources in real time; the data cleaning and annotation module is used to detect and clean abnormalities and annotate the data of the data synchronization mapping module; the data analysis and optimization module is used to identify potential correlations between data and optimize radiotherapy plans. The present invention significantly improves data consistency and accuracy by realizing standardization, synchronization and cleaning of multi-source heterogeneous data, and adopts integrated learning and genetic algorithms to optimize treatment plans, thereby promoting the personalization and precision of radiotherapy.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data processing, and in particular to a data processing system and method applied to radiotherapy. Background Art

[0002] Radiotherapy is a common cancer treatment method that uses high-energy radiation to accurately kill tumor cells while maximally protecting surrounding healthy tissues. It has been widely used in clinical practice. With the continuous development of medical imaging technology and radiotherapy equipment, the type and quantity of patient data obtained have increased significantly, including medical imaging data from different devices, patients' physiological parameter data, and various parameters during radiotherapy. The efficient integration, processing and analysis of these multi-source heterogeneous data are of great significance for formulating accurate personalized treatment plans. However, the current radiotherapy data processing system still has many technical difficulties in the integration and synchronization of multi-source data, data cleaning and annotation, and optimization of treatment plans, which restricts the further improvement of radiotherapy effects.

[0003] In the existing technology, although some radiotherapy data processing systems can process a single type of data, it is often difficult to achieve efficient integration and synchronization when faced with multi-source heterogeneous data from different devices, resulting in difficulty in ensuring data consistency and accuracy; in addition, the data cleaning and labeling process relies on manual operations, which are inefficient and error-prone, and cannot meet the needs of large-scale clinical applications; data analysis and treatment plan optimization lack intelligent support, and cannot fully explore the potential correlations in the data, affecting the accuracy and personalization of treatment plans. Summary of the invention

[0004] Based on the above objectives, the present invention provides a data processing system and method for radiotherapy.

[0005] A data processing system for radiotherapy includes a data acquisition and processing module, a data synchronization mapping module, a data cleaning and annotation module, a data analysis and optimization module, and a data visualization module; wherein;

[0006] Data acquisition and processing module: used to collect medical imaging data, patient physiological data and radiotherapy parameters from different devices, and standardize different data formats and timestamps through data conversion algorithms;

[0007] Data synchronization mapping module: used to receive standardized data from the data acquisition and processing module, and use the timing synchronization algorithm and spatial mapping algorithm to synchronize and align data from different sources in real time and space, ensuring the consistency of various types of data in time and space;

[0008] Data cleaning and annotation module: used to detect and clean the data of the data synchronization mapping module, remove noise and erroneous data, and annotate according to the patient's historical data;

[0009] Data analysis and optimization module: used to receive the cleaned and annotated data in the data cleaning and annotation module, conduct comprehensive analysis on it using an integrated learning algorithm, identify potential correlations between data, and optimize the radiotherapy plan in real time according to the progress of treatment;

[0010] Data visualization module: used to display the potential correlation between treatment plans and data from the data analysis and optimization module in an intuitive graphical manner.

[0011] Optionally, the data acquisition and processing module includes a medical image data acquisition unit, a physiological data acquisition unit, a treatment parameter acquisition unit, a format conversion unit, a time synchronization unit and a space coordinate mapping unit; wherein:

[0012] Medical imaging data acquisition unit: used to acquire three-dimensional or two-dimensional imaging data of patients from medical imaging equipment, wherein the medical imaging equipment includes CT, MRI and PET;

[0013] Physiological data acquisition unit: used to collect the patient's physiological data in real time from physiological sensors, the physiological sensors including heart rate monitors and blood pressure sensors;

[0014] Treatment parameter acquisition unit: used to collect various parameters during the treatment process from the radiotherapy equipment, including treatment dose, treatment position and radiation intensity;

[0015] Format conversion unit: used to standardize the data collected by the medical image data acquisition unit, the physiological data acquisition unit and the treatment parameter acquisition unit, and convert the different data formats of the collected medical image data, patient physiological data and radiotherapy parameters into a unified JSON data format;

[0016] Time synchronization unit: used to align the collected data from different devices according to the timestamp to ensure the consistency of the data within the same time frame;

[0017] Spatial coordinate mapping unit: used to perform coordinate conversion on spatial data collected by different devices to ensure the consistency of spatial coordinates of different data sources.

[0018] Optionally, the data synchronization mapping module includes a timing synchronization unit and a space mapping unit; wherein:

[0019] Timing synchronization unit: used to receive standardized data from the data acquisition and processing module, and synchronize data from different sources in real time according to the timestamp of each data through a timing synchronization algorithm. The timing synchronization algorithm calculates the time deviation of each data point and adjusts the timestamp of the data to ensure that data from different sources are aligned on the same time axis;

[0020] Spatial mapping unit: used to align the spatial coordinates of data from different sources using a spatial mapping algorithm based on the received standardized data, thereby eliminating the spatial deviation caused by different device positions.

[0021] Optionally, the space mapping unit specifically includes:

[0022] Coordinate transformation subunit: used to perform preliminary alignment of spatial data through a coordinate transformation algorithm according to the spatial coordinates collected by different devices. The coordinate transformation algorithm is performed through the following formula: ; ; ;in, is the coordinate in the original coordinate system, is the coordinate in the target coordinate system after transformation, is the rotation angle;

[0023] Rotation correction subunit: used to perform rotation correction on the spatial coordinates by calculating the rotation angle between the devices. The calculation formula is: ,in, and They are spatial coordinate points from different devices. is the rotation angle that needs to be corrected;

[0024] Translation correction subunit: used to perform translation correction based on the rotation-corrected data to eliminate the offset of different data sources in space. The translation correction formula is:

[0025] ;

[0026] ;

[0027] ;in, is the coordinate after translation correction, and are the spatial coordinates in and The amount of translation in the direction.

[0028] Optionally, the data cleaning and labeling module includes an anomaly detection unit, a data cleaning unit and a data labeling unit; wherein:

[0029] Anomaly detection unit: used to detect anomalies in the standardized data from the data synchronization mapping module, using a combination of statistical analysis algorithms and machine learning algorithms to operate, where the statistical analysis algorithm is used to calculate the mean of the data set and standard deviation , Application 3 The rule identifies outliers that deviate from the mean by more than three standard deviations; the machine learning algorithm uses the isolation forest algorithm to detect anomalies in the data, segmenting the data by building a random tree and calculating the anomaly score for each data point. , the formula is: , where the path length For data points The average path length in the isolation forest, The size of the dataset The expected path length of

[0030] Data cleaning unit: used to remove noise and erroneous data identified as abnormal according to the detection results of the abnormality detection unit, so as to ensure the accuracy and integrity of subsequent data;

[0031] Data annotation unit: used to annotate the cleaned data according to the patient's historical data, first receiving the patient's historical health records and treatment records ; Then, combined with the cleaned data, the preset rules are applied to annotate the data to generate annotated data sets. The specific rules include:

[0032] Rule 1: If the patient has a history record Presence of specified diseases , then the relevant physiological data Annotated as disease-related data;

[0033] Rule 2: If the treatment record There is a designated treatment plan , then the corresponding treatment parameters Mark as Scheme Related data.

[0034] Optionally, the data analysis and optimization module includes a feature extraction unit, an integrated learning unit and a treatment plan optimization unit; wherein:

[0035] Feature extraction unit: used to extract features from the annotated data set transmitted by the data cleaning and annotation module. The feature extraction unit uses the principal component analysis algorithm to reduce the dimension of high-dimensional data and retain the most informative features;

[0036] An integrated learning unit is used to receive the feature vector extracted by the feature extraction unit, and use an integrated learning algorithm to conduct a comprehensive analysis on the feature vector to identify potential correlations between data. The integrated learning algorithm includes a random forest algorithm and a gradient boosting decision tree algorithm.

[0037] Treatment plan optimization unit: It is used to optimize and adjust the radiotherapy plan based on the analysis results of the integrated learning unit and the real-time treatment progress data, and dynamically adjust the treatment plan by simulating natural selection and gene crossover.

[0038] Optionally, the integrated learning unit includes:

[0039] Random forest algorithm subunit: used to identify potential correlations between data by building multiple decision trees and voting. The specific steps include:

[0040] Step 1: Randomly sample the training data set to generate multiple training subsets;

[0041] Step 2, each training subset is used to train a decision tree model;

[0042] Step 3: Vote on the prediction results of all decision trees to finally get the classification results;

[0043] Gradient boosting decision tree algorithm subunit: used to generate multiple decision trees in an iterative manner and weight the output of each new tree to the output of the previous tree to gradually improve the prediction accuracy.

[0044] Optionally, the treatment plan optimization unit includes:

[0045] Initial population generation: Based on the analysis results provided by the integrated learning unit, multiple initial treatment plans are generated, each of which is represented by an individual ,in is the individual number, and Contains specific configuration of radiation therapy parameters;

[0046] Fitness evaluation: For each individual Calculate its fitness , the fitness function is based on the treatment effect and safety indicators To evaluate, the formula is: ,in, and are the weight coefficients of treatment effect and safety, respectively, satisfying ;

[0047] Select operation: Based on fitness The size of the individual with the highest fitness is selected proportionally Enter the breeding pool; crossover operation: randomly pair individuals from the selected breeding pool and , at a randomly selected intersection Exchange gene fragments to generate new individuals and ;

[0048] Mutation operation: for newly generated individuals and The genes undergo small random mutations.

[0049] Population update: the individuals generated after mutation and Replace individuals with low fitness in the current population to form an updated population ;

[0050] Iteration termination: Repeat the above steps until the predetermined optimization goal is met or the maximum number of iterations is reached, and finally the optimal treatment plan is output.

[0051] Optionally, the data visualization module includes a treatment plan display unit and a data association display unit; wherein:

[0052] Treatment plan display unit: used to receive the optimal treatment plan data output by the data analysis and optimization module, and use a graphical interface to display the parameters of the treatment plan, including treatment dose, treatment position and radiation intensity; the treatment plan is presented in the form of two-dimensional or three-dimensional graphics;

[0053] Data association display unit: used to display the potential association relationship between data identified in the data analysis and optimization module, and present the mutual influence between different data in the treatment process through charts or relationship diagrams; through the color change or size change of the graph, the association strength and direction between each data item are intuitively displayed.

[0054] A data processing method for radiotherapy is implemented by the above-mentioned data processing system for radiotherapy, and comprises the following steps:

[0055] S1: Collect medical imaging data, patient physiological data and radiotherapy parameters from different devices;

[0056] S2: Standardize the data format and timestamp collected by S1 through data conversion algorithms;

[0057] S3: Use the time synchronization algorithm and spatial mapping algorithm to synchronize and align the data standardized by S2 in real time;

[0058] S4: Perform anomaly detection and cleaning on the synchronized and aligned data, remove noise and erroneous data, and annotate the data based on the patient's historical data;

[0059] S5: Use ensemble learning algorithms to comprehensively analyze the cleaned and labeled data, identify potential correlations between data, and optimize radiotherapy plans in real time based on the analysis results.

[0060] Beneficial effects of the present invention:

[0061] The present invention significantly improves the consistency and accuracy of data by standardizing, synchronizing and spatially aligning medical imaging data, patient physiological data and radiotherapy parameter data from different devices; by efficiently cleaning and labeling the data, noise and erroneous data are automatically removed, thus avoiding errors that may occur in manual operations and ensuring the reliability and efficiency of data processing.

[0062] The present invention uses an integrated learning algorithm and a genetic algorithm to perform comprehensive analysis on the processed data and optimize the treatment plan, thereby being able to explore the potential correlation between the data, dynamically adjust the radiotherapy plan, and achieve personalized and precise treatment; compared with traditional treatment plan optimization methods, the method of the present invention can better combine real-time treatment progress data, optimize the radiotherapy plan, and thus significantly improve the treatment effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions in the present invention or 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 in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0064] Figure 1 A schematic diagram of a data processing system for radiotherapy according to an embodiment of the present invention;

[0065] Figure 2 Schematic diagram of a data processing method for radiotherapy according to an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0067] like Figure 1As shown, a data processing system applied to radiotherapy includes a data acquisition and processing module, a data synchronization mapping module, a data cleaning and annotation module, a data analysis and optimization module, and a data visualization module; wherein;

[0068] Data acquisition and processing module: used to collect medical imaging data, patient physiological data and radiotherapy parameters from different devices, and standardize different data formats and timestamps through data conversion algorithms;

[0069] Data synchronization mapping module: used to receive standardized data from the data acquisition and processing module, and use the timing synchronization algorithm and spatial mapping algorithm to synchronize and align data from different sources in real time and space, ensuring the consistency of various types of data in time and space;

[0070] Data cleaning and annotation module: used to detect and clean the data of the data synchronization mapping module, remove noise and erroneous data, and annotate according to the patient's historical data;

[0071] Data analysis and optimization module: used to receive the cleaned and annotated data in the data cleaning and annotation module, conduct comprehensive analysis on it using an integrated learning algorithm, identify potential correlations between data, and optimize the radiotherapy plan in real time according to the progress of treatment;

[0072] Data visualization module: used to display the potential correlation between treatment plans and data from the data analysis and optimization module in an intuitive graphical manner.

[0073] Optionally, the data acquisition and processing module includes a medical image data acquisition unit, a physiological data acquisition unit, a treatment parameter acquisition unit, a format conversion unit, a time synchronization unit and a space coordinate mapping unit; wherein:

[0074] Medical imaging data acquisition unit: used to acquire three-dimensional or two-dimensional imaging data of patients from medical imaging equipment, wherein the medical imaging equipment includes CT, MRI and PET;

[0075] Physiological data acquisition unit: used to collect the patient's physiological data in real time from physiological sensors, the physiological sensors including heart rate monitors and blood pressure sensors;

[0076] Treatment parameter acquisition unit: used to collect various parameters during the treatment process from the radiotherapy equipment, including treatment dose, treatment position and radiation intensity;

[0077] Format conversion unit: used to standardize the data collected by the medical image data acquisition unit, the physiological data acquisition unit and the treatment parameter acquisition unit, and convert the different data formats of the collected medical image data, patient physiological data and radiotherapy parameters into a unified JSON data format;

[0078] Time synchronization unit: used to align the collected data from different devices according to the timestamp to ensure the consistency of the data within the same time frame;

[0079] Spatial coordinate mapping unit: used to perform coordinate conversion on spatial data collected by different devices to ensure the consistency of spatial coordinates of different data sources.

[0080] Optionally, the data synchronization mapping module includes a timing synchronization unit and a space mapping unit; wherein:

[0081] Timing synchronization unit: used to receive standardized data from the data acquisition and processing module, and synchronize data from different sources in real time according to the timestamp of each data through a timing synchronization algorithm. The timing synchronization algorithm calculates the time deviation of each data point and adjusts the timestamp of the data to ensure that data from different sources are aligned on the same time axis;

[0082] Spatial mapping unit: used to align the spatial coordinates of data from different sources using a spatial mapping algorithm based on the received standardized data, thereby eliminating the spatial deviation caused by different device positions.

[0083] Optionally, the space mapping unit specifically includes:

[0084] Coordinate transformation subunit: used to perform preliminary alignment of spatial data through a coordinate transformation algorithm according to the spatial coordinates collected by different devices. The coordinate transformation algorithm is performed through the following formula: ; ; ;in, is the coordinate in the original coordinate system, is the coordinate in the target coordinate system after transformation, is the rotation angle;

[0085] Rotation correction subunit: used to perform rotation correction on the spatial coordinates by calculating the rotation angle between the devices. The calculation formula is: ,in, and They are spatial coordinate points from different devices. is the rotation angle that needs to be corrected;

[0086] Translation correction subunit: used to perform translation correction based on the rotation-corrected data to eliminate the offset of different data sources in space. The translation correction formula is:

[0087] ;

[0088] ;

[0089] ;in, is the coordinate after translation correction, and are the spatial coordinates in and The amount of translation in the direction.

[0090] Optionally, the data cleaning and labeling module includes an anomaly detection unit, a data cleaning unit and a data labeling unit; wherein:

[0091] Anomaly detection unit: used to detect anomalies in the standardized data from the data synchronization mapping module, using a combination of statistical analysis algorithms and machine learning algorithms to operate, where the statistical analysis algorithm is used to calculate the mean of the data set and standard deviation , Application 3 The rule identifies outliers that deviate from the mean by more than three standard deviations; the machine learning algorithm uses the isolation forest algorithm to detect anomalies in the data, segmenting the data by building a random tree and calculating the anomaly score for each data point. , the formula is: , where the path length For data points The average path length in the isolation forest, The size of the dataset The expected path length of

[0092] Data cleaning unit: used to remove noise and erroneous data identified as abnormal according to the detection results of the abnormality detection unit, so as to ensure the accuracy and integrity of subsequent data;

[0093] Data annotation unit: used to annotate the cleaned data according to the patient's historical data, first receiving the patient's historical health records and treatment records ; Then, combined with the cleaned data, the preset rules are applied to annotate the data to generate annotated data sets. The specific rules include:

[0094] Rule 1: If the patient has a history record Presence of specified diseases , then the relevant physiological data Annotated as disease-related data;

[0095] Rule 2: If the treatment record There is a designated treatment plan , then the corresponding treatment parameters Mark as Scheme Related data.

[0096] Optionally, the data analysis and optimization module includes a feature extraction unit, an integrated learning unit and a treatment plan optimization unit; wherein:

[0097] Feature extraction unit: used to extract features from the annotated data set transmitted by the data cleaning and annotation module. The feature extraction unit uses the principal component analysis algorithm to reduce the dimension of high-dimensional data and retain the most informative features;

[0098] An integrated learning unit is used to receive the feature vector extracted by the feature extraction unit, and use an integrated learning algorithm to conduct a comprehensive analysis on the feature vector to identify potential correlations between data. The integrated learning algorithm includes a random forest algorithm and a gradient boosting decision tree algorithm.

[0099] Treatment plan optimization unit: It is used to optimize and adjust the radiotherapy plan based on the analysis results of the integrated learning unit and the real-time treatment progress data, and dynamically adjust the treatment plan by simulating natural selection and gene crossover.

[0100] Optionally, the integrated learning unit includes:

[0101] Random forest algorithm subunit: used to identify potential correlations between data by building multiple decision trees and voting. The specific steps include:

[0102] Step 1: Randomly sample the training data set to generate multiple training subsets;

[0103] Step 2, each training subset is used to train a decision tree model;

[0104] Step 3: Vote on the prediction results of all decision trees to finally get the classification results;

[0105] Gradient boosting decision tree algorithm subunit: used to generate multiple decision trees in an iterative manner and weight the output of each new tree to the output of the previous tree to gradually improve the prediction accuracy.

[0106] Optionally, the treatment plan optimization unit includes:

[0107] Initial population generation: Based on the analysis results provided by the integrated learning unit, multiple initial treatment plans are generated, each of which is represented by an individual ,in is the individual number, and Contains specific configuration of radiation therapy parameters;

[0108] Fitness evaluation: For each individual Calculate its fitness , the fitness function is based on the treatment effect and safety indicators To evaluate, the formula is: ,in, and are the weight coefficients of treatment effect and safety, respectively, satisfying ;

[0109] Select operation: Based on fitness The size of the individual with the highest fitness is selected proportionally Enter the breeding pool; crossover operation: randomly pair individuals from the selected breeding pool and , at a randomly selected intersection Exchange gene fragments to generate new individuals and ;

[0110] Mutation operation: for newly generated individuals and The genes undergo small random mutations.

[0111] Population update: the individuals generated after mutation and Replace individuals with low fitness in the current population to form an updated population ;

[0112] Iteration termination: Repeat the above steps until the predetermined optimization goal is met or the maximum number of iterations is reached, and finally the optimal treatment plan is output.

[0113] Optionally, the data visualization module includes a treatment plan display unit and a data association display unit; wherein:

[0114] Treatment plan display unit: used to receive the optimal treatment plan data output by the data analysis and optimization module, and use a graphical interface to display the parameters of the treatment plan, including treatment dose, treatment position and radiation intensity; the treatment plan is presented in the form of two-dimensional or three-dimensional graphics;

[0115] Data association display unit: used to display the potential association relationship between data identified in the data analysis and optimization module, and present the mutual influence between different data in the treatment process through charts or relationship diagrams; through the color change or size change of the graph, the association strength and direction between each data item are intuitively displayed.

[0116] like Figure 2 As shown, a data processing method for radiotherapy is implemented by the above-mentioned data processing system for radiotherapy, comprising the following steps:

[0117] S1: Collect medical imaging data, patient physiological data and radiotherapy parameters from different devices;

[0118] S2: Standardize the data format and timestamp collected by S1 through data conversion algorithms;

[0119] S3: Use the time synchronization algorithm and spatial mapping algorithm to synchronize and align the data standardized by S2 in real time;

[0120] S4: Perform anomaly detection and cleaning on the synchronized and aligned data, remove noise and erroneous data, and annotate the data based on the patient's historical data;

[0121] S5: Use ensemble learning algorithms to comprehensively analyze the cleaned and labeled data, identify potential correlations between data, and optimize radiotherapy plans in real time based on the analysis results.

[0122] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0123] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A data processing system for radiotherapy, characterized in that: It includes data acquisition and processing module, data synchronization mapping module, data cleaning and annotation module, data analysis and optimization module and data visualization module; in; Data acquisition and processing module: used to collect medical imaging data, patient physiological data and radiotherapy parameters from different devices, and standardize different data formats and timestamps through data conversion algorithms; Data synchronization mapping module: used to receive standardized data from the data acquisition and processing module, and use the timing synchronization algorithm and spatial mapping algorithm to synchronize and align data from different sources in real time and space, ensuring the consistency of various types of data in time and space; Data cleaning and annotation module: used to detect and clean the data of the data synchronization mapping module, remove noise and erroneous data, and annotate according to the patient's historical data; Data analysis and optimization module: used to receive the cleaned and annotated data in the data cleaning and annotation module, conduct comprehensive analysis on it using an integrated learning algorithm, identify potential correlations between data, and optimize the radiotherapy plan in real time according to the progress of treatment; Data visualization module: used to display the potential relationship between the treatment plan and data from the data analysis and optimization module in an intuitive graphical way; The data acquisition and processing module includes a medical image data acquisition unit, a physiological data acquisition unit, a treatment parameter acquisition unit, a format conversion unit, a time synchronization unit and a space coordinate mapping unit; wherein: Medical imaging data acquisition unit: used to acquire three-dimensional or two-dimensional imaging data of patients from medical imaging equipment, wherein the medical imaging equipment includes CT, MRI and PET; Physiological data acquisition unit: used to collect the patient's physiological data in real time from physiological sensors, the physiological sensors including heart rate monitors and blood pressure sensors; Treatment parameter acquisition unit: used to collect various parameters during the treatment process from the radiotherapy equipment, including treatment dose, treatment position and radiation intensity; Format conversion unit: used to standardize the data collected by the medical image data acquisition unit, the physiological data acquisition unit and the treatment parameter acquisition unit, and convert the different data formats of the collected medical image data, patient physiological data and radiotherapy parameters into a unified JSON data format; Time synchronization unit: used to align the collected data from different devices according to the timestamp to ensure the consistency of the data within the same time frame; Spatial coordinate mapping unit: used to convert the spatial data collected by different devices to ensure the consistency of spatial coordinates of different data sources; The data cleaning and annotation module includes an anomaly detection unit, a data cleaning unit and a data annotation unit; wherein: Anomaly detection unit: used to detect anomalies in the standardized data from the data synchronization mapping module, using a combination of statistical analysis algorithms and machine learning algorithms to operate, where the statistical analysis algorithm is used to calculate the mean of the data set and standard deviation , Application 3 The rule identifies outliers that deviate from the mean by more than three standard deviations; the machine learning algorithm uses the isolation forest algorithm to detect anomalies in the data, segmenting the data by building a random tree and calculating the anomaly score for each data point. , the formula is: , where the path length For data points The average path length in the isolation forest, The size of the dataset The expected path length of Data cleaning unit: used to remove noise and erroneous data identified as abnormal according to the detection results of the abnormality detection unit, so as to ensure the accuracy and integrity of subsequent data; Data annotation unit: used to annotate the cleaned data according to the patient's historical data, first receiving the patient's historical health records and treatment records ; Then, combined with the cleaned data, the preset rules are applied to annotate the data to generate annotated data sets. The specific rules include: Rule 1: If the patient has a history record Presence of specified diseases , then the relevant physiological data Annotated as disease-related data; Rule 2: If the treatment record There is a designated treatment plan , then the corresponding treatment parameters Mark as Scheme relevant data; The data analysis and optimization module includes a feature extraction unit, an integrated learning unit and a treatment plan optimization unit; wherein: Feature extraction unit: used to extract features from the annotated data set transmitted by the data cleaning and annotation module. The feature extraction unit uses the principal component analysis algorithm to reduce the dimension of high-dimensional data and retain the most informative features; An integrated learning unit is used to receive the feature vector extracted by the feature extraction unit, and use an integrated learning algorithm to conduct a comprehensive analysis on the feature vector to identify potential correlations between data. The integrated learning algorithm includes a random forest algorithm and a gradient boosting decision tree algorithm. Treatment plan optimization unit: It is used to optimize and adjust the radiotherapy plan based on the analysis results of the integrated learning unit and the real-time treatment progress data, and dynamically adjust the treatment plan by simulating natural selection and gene crossover.

2. A data processing system for radiotherapy according to claim 1, characterized in that: The data synchronization mapping module includes a timing synchronization unit and a space mapping unit; wherein: Timing synchronization unit: used to receive standardized data from the data acquisition and processing module, and synchronize data from different sources in real time according to the timestamp of each data through a timing synchronization algorithm. The timing synchronization algorithm calculates the time deviation of each data point and adjusts the timestamp of the data to ensure that data from different sources are aligned on the same time axis; Spatial mapping unit: used to align the spatial coordinates of data from different sources using a spatial mapping algorithm based on the received standardized data, thereby eliminating the spatial deviation caused by different device positions.

3. A data processing system for radiotherapy according to claim 2, characterized in that: The space mapping unit specifically includes: Coordinate transformation subunit: used to perform preliminary alignment of spatial data through a coordinate transformation algorithm according to the spatial coordinates collected by different devices. The coordinate transformation algorithm is performed through the following formula: ; ; ;in, is the coordinate in the original coordinate system, is the coordinate in the target coordinate system after transformation, is the rotation angle; Rotation correction subunit: used to perform rotation correction on the spatial coordinates by calculating the rotation angle between the devices. The calculation formula is: ,in, and They are spatial coordinate points from different devices. is the rotation angle that needs to be corrected; Translation correction subunit: used to perform translation correction based on the rotation-corrected data to eliminate the offset of different data sources in space. The translation correction formula is: ； ； ;in, is the coordinate after translation correction, and are the spatial coordinates in and The amount of translation in the direction.

4. A data processing system for radiotherapy according to claim 1, characterized in that: The integrated learning unit comprises: Random forest algorithm subunit: used to identify potential correlations between data by building multiple decision trees and voting. The specific steps include: Step 1: Randomly sample the training data set to generate multiple training subsets; Step 2, each training subset is used to train a decision tree model; Step 3: Vote on the prediction results of all decision trees to finally get the classification results; Gradient boosting decision tree algorithm subunit: used to generate multiple decision trees in an iterative manner and weight the output of each new tree to the output of the previous tree to gradually improve the prediction accuracy.

5. A data processing system for radiotherapy according to claim 4, characterized in that: The treatment plan optimization unit comprises: Initial population generation: Based on the analysis results provided by the integrated learning unit, multiple initial treatment plans are generated, each of which is represented by an individual ,in is the individual number, and Contains specific configuration of radiation therapy parameters; Fitness evaluation: For each individual Calculate its fitness , the fitness function is based on the treatment effect and safety indicators To evaluate, the formula is: ,in, and are the weight coefficients of treatment effect and safety, respectively, satisfying ; Select operation: Based on fitness The size of the individual with the highest fitness is selected proportionally Enter the breeding pond; Crossover operation: randomly pairing individuals from a selected breeding pool and , at a randomly selected intersection Exchange gene fragments to generate new individuals and ; Mutation operation: for newly generated individuals and The genes undergo small random mutations. Population update: The individuals generated after mutation and Replace individuals with low fitness in the current population to form an updated population ; Iteration termination: Repeat the above steps until the predetermined optimization goal is met or the maximum number of iterations is reached, and finally the optimal treatment plan is output.

6. A data processing system for radiotherapy according to claim 1, characterized in that: The data visualization module includes a treatment plan display unit and a data association display unit; wherein: Treatment plan display unit: used to receive the optimal treatment plan data output by the data analysis and optimization module, and use a graphical interface to display the parameters of the treatment plan, including treatment dose, treatment position and radiation intensity; the treatment plan is presented in the form of two-dimensional or three-dimensional graphics; Data association display unit: used to display the potential association relationship between data identified in the data analysis and optimization module, and present the mutual influence between different data in the treatment process through charts or relationship diagrams; through the color change or size change of the graph, the association strength and direction between each data item are intuitively displayed.

7. A data processing method for radiotherapy, implemented by a data processing system for radiotherapy according to any one of claims 1 to 6, characterized in that: The following steps are involved: S1: Collect medical imaging data, patient physiological data and radiotherapy parameters from different devices; S2: Standardize the data format and timestamp collected by S1 through data conversion algorithms; S3: Use the time synchronization algorithm and spatial mapping algorithm to synchronize and align the data standardized by S2 in real time; S4: Perform anomaly detection and cleaning on the synchronized and aligned data, remove noise and erroneous data, and annotate the data based on the patient's historical data; S5: Use ensemble learning algorithms to comprehensively analyze the cleaned and labeled data, identify potential correlations between data, and optimize radiotherapy plans in real time based on the analysis results.

Images