Radiotherapy dose real-time monitoring system

Abstract

The invention provides a radiotherapy dose real-time monitoring system, which comprises a data interface module for acquiring multi-source data related to radiotherapy in real time, the multi-source data comprises plan data imported from a treatment plan system, and the plan data comprises three-dimensional plan dose distribution and plan anatomical structure data; according to the invention, the real-time dynamic monitoring of the radiation treatment dose is realized, the limitation that only pre-verification before treatment or offline analysis after treatment can be carried out in the traditional technology is broken through, the treatment plan data, the machine operation parameters and the real-time anatomical image of the patient are integrated, and the real-time monitoring of the radiation treatment dose is realized. A multi-source data fusion processing mechanism is innovatively constructed, the comprehensive influence of machine execution deviation and anatomical change of a patient on the dose is quantified in real time, actual dose distribution in a body is dynamically reconstructed, dose delivery is changed from an invisible black box state to a real-time traceable visual process, and accurate monitoring of the radiotherapy dose in treatment is truly realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical monitoring, in particular to a radiotherapy dose real-time monitoring system. BACKGROUND

[0002] Radiotherapy is a cancer treatment method that uses high-energy radiation such as X-rays or proton beams to target and destroy cancer cells, aiming to inhibit tumor growth or eliminate malignant cells. During radiotherapy, dose monitoring is crucial as it ensures that the radiation dose is accurately delivered to the tumor site while minimizing damage to surrounding healthy tissues. The reason is that radiation has a cumulative effect: too high a dose can cause damage to normal tissues, leading to side effects such as skin burns, organ dysfunction, and even secondary cancers; while too low a dose may not effectively kill cancer cells, increasing the risk of recurrence. Through real-time monitoring, the medical team can verify whether the treatment plan is accurately implemented, adjust equipment errors in a timely manner, and optimize the plan based on individual responses. Evidence shows that strict dose control can significantly improve treatment success rates, reduce complications, and thus improve patient survival rates and quality of life, so dose monitoring is a core link to ensure the safety and effectiveness of radiotherapy.

[0003] Existing radiotherapy dose monitoring techniques have significant defects, mainly relying on pre-treatment phantom verification or post-treatment offline analysis, and cannot achieve real-time dynamic monitoring during treatment. For example, conventional pre-verification using ionization chamber arrays or films can only simulate ideal conditions and cannot capture the impact of machine execution deviations (such as multi-leaf collimator leaf position errors or gantry angle shifts) on the dose during treatment; while dose verification based on electronic portal imaging devices (EPID) usually only reviews data after irradiation, which cannot intervene in real time. Clinical cases show that in lung cancer radiotherapy, respiratory motion causes tumor target displacement, or in prostate cancer treatment, organ movement is caused by rectal filling changes. Traditional systems lack real-time image fusion mechanisms and cannot correct dose deviations caused by anatomical changes, resulting in discrepancies between actual irradiation dose and plan (deviation can be more than 5%), and lack real-time warning functions, so the error irradiation may continue throughout the treatment, increasing the risk of damage to normal tissues. At the same time, isolated processing of multi-source data (plans, machine parameters, and images) causes verification lag and incompleteness, making quality traceability dependent on manual labor, which is inefficient. SUMMARY

[0004] TECHNICAL PROBLEM SOLVED In view of the deficiencies of the prior art, the present application provides a radiotherapy dose real-time monitoring system, which solves the problem that the existing radiotherapy dose monitoring techniques mainly rely on pre-treatment phantom verification or post-treatment offline analysis, and cannot achieve real-time dynamic monitoring during treatment.

[0005] TECHNICAL SCHEME To achieve the above object, the present application is implemented by the following technical solutions: A radiotherapy dose real-time monitoring system, comprising A data interface module, which acquires radiotherapy-related multi-source data in real time, wherein the multi-source data includes: plan data imported from a treatment planning system, the plan data containing three-dimensional plan dose distribution and plan anatomic structure data; machine operation parameters collected from a radiotherapy machine in real time; patient real-time anatomic image data collected from an image-guided device in real time; A data processing and fusion module, which is connected with the data interface module, analyzes the machine operation parameters to obtain actual irradiation parameters; performs morphing registration on the patient real-time anatomic image data and the plan anatomic structure data to obtain change information of patient anatomic structure; and based on the actual irradiation parameters and the change information, generates three-dimensional actual dose distribution in the patient's body by real-time correction on the three-dimensional plan dose distribution through a dose reconstruction algorithm. A dose comparison and early warning module, which is connected with the data processing and fusion module, compares the three-dimensional actual dose distribution with the three-dimensional plan dose distribution at a voxel level to generate dose deviation quantization results; and is configured to trigger an early warning signal when the dose deviation quantization results exceed a preset safety threshold.

[0006] Preferably, the data interface module comprises: A plan data receiving unit, which receives the plan data from a TPS based on a DICOMRT protocol; A machine parameter receiving unit, which reads a log file from a control system of a radiotherapy machine in real time based on a network communication protocol, wherein the log file contains multileaf collimator leaf position, gantry angle, collimator angle and monitor unit value; An image data receiving unit, which receives two-dimensional or three-dimensional patient real-time anatomic image data from a KV-CBCT, a MV-CT or an electronic portal imaging device.

[0007] Preferably, the data processing and fusion module comprises: A parameter analysis unit, which is used to extract MLC leaf actual position and gantry actual angle from the machine operation parameters and calculate the deviation thereof from the planned values; An anatomic registration unit, which is used to match real-time image and plan CT image through a morphing image registration algorithm to output a morphing field representing organ position and density change; A dose reconstruction unit, which is used to calculate the three-dimensional actual dose distribution by a pre-calculated dose perturbation algorithm or a GPU-accelerated fast Monte Carlo algorithm, with the three-dimensional plan dose distribution as a reference, in combination with the deviation and the morphing field.

[0008] Preferably, the pre-calculated dose perturbation algorithm is: calculating a reference dose cloud by high-precision algorithm based on planning parameters in advance; during treatment, performing linear or nonlinear perturbation on the reference dose cloud according to real-time machine parameter deviation and anatomical changes to quickly output the corrected dose distribution.

[0009] Preferably, the dose comparison and early warning module comprises: a dose comparison unit for calculating the absolute dose difference and the relative dose deviation between the three-dimensional actual dose distribution and the three-dimensional planning dose distribution; a gamma analysis unit for performing gamma analysis based on the pass rate index to evaluate the coincidence of the dose distribution in the spatial and dose dimensions; a warning unit configured to activate an audio-visual alarm device and generate a suggestion prompt to pause treatment when the relative dose deviation exceeds 5% or the gamma pass rate is lower than 95%.

[0010] Preferably, it further comprises a data storage and report generation module connected with the data interface module, the data processing and fusion module, and the dose comparison and early warning module, for structured storage of all real-time data, intermediate processing results and final comparison results, and automatic generation of structured reports conforming to clinical quality assurance specifications.

[0011] Advantages The present application provides a radiotherapy dose real-time monitoring system. It has the following advantages: The present application realizes real-time dynamic monitoring of radiotherapy dose, breaks through the limitation of traditional technology which can only perform pre-treatment verification or offline analysis after treatment, integrates treatment planning data, machine operating parameters and patient real-time anatomical images, innovatively constructs a multi-source data fusion processing mechanism, quantifies the comprehensive influence of machine execution deviation and patient anatomical changes on dose in real time, dynamically reconstructs the actual dose distribution in the body, changes the dose delivery from an invisible black box state to a real-time traceable visual process, and truly realizes the accurate monitoring of radiotherapy dose during treatment, filling the gap in dynamic dose verification in the field.

[0012] Through real-time dose comparison and active early warning mechanism, the present application can timely identify clinically unacceptable dose deviation and trigger intervention prompt, avoiding the risk of continuous irradiation of incorrect dose. At the same time, complete data storage and standardized report function provide reliable basis for radiotherapy quality traceability, simplify clinical quality control process, provide direct support for treatment plan optimization and department quality review, and effectively help the standardized implementation of precise radiotherapy, reducing the risk of treatment-related complications. BRIEF DESCRIPTION OF DRAWINGS

[0013] Figure 1 The present application provides a flowchart of a radiotherapy dose real-time monitoring system. Detailed Implementation

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] Example: like Figure 1 As shown, this embodiment of the invention provides a real-time radiotherapy dose monitoring system, including... The data interface module acquires multi-source data related to radiotherapy in real time. This multi-source data includes: treatment planning data imported from the treatment planning system, which contains three-dimensional planned dose distribution and planned anatomical structure data; machine operating parameters acquired in real time from the radiotherapy machine; and real-time anatomical image data of the patient acquired in real time from the image-guided equipment. The core function of this module is to break down data barriers between different devices in the radiotherapy process, enabling multi-dimensional, real-time data acquisition. Specifically: The "planning data" obtained from the treatment planning system (TPS) serves as a "benchmark reference" for monitoring. It includes three-dimensional planned dose distribution (such as the target area prescription dose and the organ at risk limit dose) and planned anatomical data (such as the contours of the tumor, spinal cord, and other tissues in the planned CT scan), providing a standard for subsequent comparisons. The "machine operating parameters" collected from the radiotherapy machine are "dynamic variables" that reflect the dose delivery process. These include the position of the multileaf grating (MLC) and the gantry angle, which directly affect the shape and incident direction of the radiation beam and are key factors that cause the deviation between the actual dose and the planned dose. The "real-time anatomical imaging data of the patient" collected from the image-guided equipment is a "real-time feedback" reflecting the patient's condition. For example, tumor displacement or organ deformation shown by KV-CBCT during treatment can alter the attenuation and scattering of radiation in the body, thereby affecting the actual absorbed dose.

[0016] The data processing and fusion module connects to the data interface module, analyzes machine operating parameters to obtain actual irradiation parameters, performs deformation registration between real-time anatomical image data of the patient and planned anatomical structure data to obtain information on changes in the patient's anatomical structure, and, based on the actual irradiation parameters and changes, performs real-time correction of the three-dimensional planned dose distribution through a dose reconstruction algorithm to generate the three-dimensional actual dose distribution in the patient's body. This module serves as the system's "computational hub." By processing and fusing multi-source data, it transforms the raw data into a "three-dimensional actual dose distribution" that can be used for comparison, thus solving the core problem of "how to quantify the actual absorbed dose in a patient's body in real time." The machine operation parameters are analyzed to obtain the "actual irradiation parameters", and the focus is on extracting the differences from the planned parameters (such as the deviation of the actual position of the MLC leaf from the planned position, the shift of the gantry rotation angle), which will directly change the coverage and shape of the ray beam; The deformed registration of the patient's real-time anatomic image and the planned anatomic structure obtains the "anatomic structure change information" (such as the moving direction and distance of the tumor, the change of tissue density), which will cause the change of the transmission path of the rays in the body (such as the change of the lung inflation state will affect the attenuation coefficient of the rays); Based on the actual irradiation parameters and the anatomic change information, the three-dimensional planned dose distribution is corrected by a dose reconstruction algorithm, and finally the "three-dimensional actual dose distribution" is generated. The core of this process is to quantify the influence of "machine execution error" and "patient state change" on the dose, to ensure that the reconstructed actual dose can truly reflect the current irradiation condition of the patient.

[0017] The dose comparison and early warning module, connected with the data processing and fusion module, compares the three-dimensional actual dose distribution with the three-dimensional planned dose distribution at the voxel level, generates dose deviation quantization results, and is configured to trigger a warning signal when the dose deviation quantization results exceed the preset safety threshold.

[0018] This module realizes "accurate comparison of actual dose and planned dose" and "timely intervention of abnormal deviation", and solves the problems of "feedback lag" and "inability to actively warn" in the prior art: Voxel-level comparison is the key to ensuring accuracy. By comparing the three-dimensional actual dose with the planned dose point by point, the specific deviation area can be located (such as insufficient dose in a part of the target area, excessive dose in a certain area of the organ at risk); The dose deviation quantization results (such as local dose deviation value, overall coincidence index) provide objective basis for clinical judgment; The preset safety threshold (such as the upper limit of dose deviation based on clinical experience, the upper limit of spatial position deviation) ensures that the system can trigger a warning immediately when the deviation exceeds the clinically acceptable range, prompting the operator to pause the treatment and investigate the cause (such as machine failure, patient position change), to avoid the continuous delivery of incorrect dose.

[0019] The data interface module includes: The planned data receiving unit receives the planned data from the TPS based on the DICOMRT protocol; Machine parameter receiving unit, based on network communication protocol, real-time read log file from the control system of the radiotherapy machine, the log file contains multileaf collimator leaf position, gantry angle, collimator angle and monitor unit value, based on network communication protocol, read log file from the control system of the radiotherapy machine, focus on obtaining MLC leaf position, gantry angle, collimator angle, monitor unit and other core parameters. Real-time reading of log file ensures that dynamic changes in machine operation can be captured; Image data receiving unit, receiving two-dimensional or three-dimensional real-time anatomical image data of patients from KV-CBCT, MV-CT or electronic portal imaging device, compatible with various image-guided devices, can receive two-dimensional or three-dimensional real-time images. These images are not only used for anatomical structure change analysis, but also assist in verifying patient positioning accuracy, providing more comprehensive anatomical basis for dose reconstruction.

[0020] Data processing and fusion module includes: Parameter analysis unit, used to extract MLC leaf actual position, gantry actual angle from machine operation parameters, and calculate the deviation from the planned value; Anatomical registration unit, used to match real-time images with planning CT images through deformation image registration algorithm, output deformation field representing organ position and density changes, match real-time images with planning CT through deformation image registration algorithm, output "deformation field". For example, respiration causes liver tumor to move 5mm to the head side, deformation field can quantify this displacement, while reflecting the density changes of the surrounding tissue of the tumor; Dose reconstruction unit, used to calculate three-dimensional actual dose distribution based on three-dimensional planned dose distribution, combined with deviation and deformation field, through pre-calculated dose perturbation algorithm or GPU accelerated fast Monte Carlo algorithm; based on three-dimensional planned dose distribution, combined with parameter deviation and deformation field to correct the dose. Adopting "pre-calculated dose perturbation algorithm" or "GPU accelerated fast Monte Carlo algorithm" to balance accuracy and speed: the former is suitable for scenarios with extremely high computing speed requirements (such as dynamic radiotherapy), and the latter is suitable for scenarios with higher accuracy requirements (such as proton therapy). The combination of the two algorithms ensures that the system can meet the needs of different radiotherapy techniques.

[0021] Pre-calculated dose perturbation algorithm: based on the planned parameters, calculate the baseline dose cloud through high-precision algorithm in advance; during treatment, linear or nonlinear perturbation is performed on the baseline dose cloud according to the real-time machine parameter deviation and anatomical changes to quickly output the corrected dose distribution; Offline stage: based on TPS planning parameters, calculate the baseline dose cloud through high-precision algorithm and store it in the system memory. Although this step is time-consuming, it can be completed before treatment and does not affect real-time performance; Online stage: During treatment, real-time machine parameter bias and anatomical changes are acquired, and the baseline dose cloud is quickly adjusted through pre-set mathematical models (e.g. linear correction model, non-linear perturbation function). For example, if MLC bias causes the shape of the field to change, the dose in the corresponding area can be corrected by adjusting the field weight formula; if the anatomical changes cause the tissue density to decrease, the dose deposition of the rays in that area can be adjusted by the attenuation coefficient correction formula.

[0022] The dose comparison and early warning module includes: A dose comparison unit for calculating the absolute dose difference and relative dose deviation between the three-dimensional actual dose distribution and the three-dimensional planned dose distribution; A gamma analysis unit for performing gamma analysis based on pass rate indicators to evaluate the coincidence of the dose distribution in the spatial and dose dimensions; An early warning unit configured to activate audio-visual alarm devices and generate a suggestion prompt to pause treatment when the relative dose deviation exceeds 5% or the gamma pass rate is less than 95%, triggering an early warning based on clinical safety thresholds. Setting "relative dose deviation exceeds 5%" or "gamma pass rate is less than 95%" as the threshold is because: ① A dose deviation of 5% can significantly affect tumor control rate or normal tissue complication risk; ② 95% of the gamma pass rate is the clinically accepted lower limit. Early warning actions include activating audio-visual alarm devices (such as console red light flashing, beeping prompt) and generating a suggestion to pause treatment, ensuring that the operator can respond quickly and intervene.

[0023] It also includes a data storage and report generation module connected to the data interface module, data processing and fusion module, and dose comparison and early warning module, for structured storage of all real-time data, intermediate processing results, and final comparison results, and automatic generation of structured reports that meet clinical quality assurance specifications; The structured storage of full-process data includes: ① raw data (TPS plan, machine log, real-time image); ② intermediate processing results (parameter bias, deformation field, reconstructed dose); ③ final comparison results (dose deviation, gamma pass rate, early warning record). Data storage uses encrypted format, in line with medical data privacy standards (such as HIPAA), and can trace the dose monitoring details of each treatment.

[0024] Report generation function: automatically generate structured reports that meet clinical quality assurance (QA) specifications, including: ① treatment overview (patient information, treatment fraction, irradiation field parameters); ② dose comparison results (dose deviation statistics of target area / organ at risk, gamma pass rate heat map); ③ early warning event record (time, deviation value, handling measures); ④ summary conclusion (e.g. "this treatment dose deviation is within acceptable range" or "MLC bias triggered early warning, treatment has been paused"). This report provides an objective basis for radiotherapy quality control, which can be used for department QA review, treatment plan optimization, and medical dispute tracing While embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely exemplary of the principles and application of the present application. Numerous modifications and adaptions can be made without departing from the spirit and scope of the present application, which is defined by the following claims and their equivalents.

Claims

1. A real-time radiotherapy dose monitoring system, characterized in that, include The data interface module acquires multi-source data related to radiotherapy in real time. The multi-source data includes: planning data imported from the treatment planning system, which includes three-dimensional planned dose distribution and planned anatomical structure data; machine operating parameters collected in real time from the radiotherapy machine; and real-time anatomical image data of the patient collected in real time from the image-guided device. The data processing and fusion module is connected to the data interface module, which parses the machine operating parameters to obtain the actual irradiation parameters; performs deformation registration between the real-time anatomical image data of the patient and the planned anatomical structure data to obtain the change information of the patient's anatomical structure; and, based on the actual irradiation parameters and the change information, performs real-time correction of the three-dimensional planned dose distribution through a dose reconstruction algorithm to generate the three-dimensional actual dose distribution in the patient's body. The dose comparison and early warning module is connected to the data processing and fusion module. It performs a voxel-level comparison between the actual three-dimensional dose distribution and the planned three-dimensional dose distribution to generate a dose deviation quantification result. It is configured to trigger an early warning signal when the dose deviation quantification result exceeds a preset safety threshold.

2. The real-time radiotherapy dose monitoring system according to claim 1, characterized in that: The data interface module includes: The planning data receiving unit receives the planning data from the TPS based on the DICOMRT protocol; The machine parameter receiving unit reads log files from the control system of the radiotherapy machine in real time based on a network communication protocol. The log files include the position of the multi-leaf grating blades, the gantry angle, the collimator angle, and the monitor unit value. The image data receiving unit receives two-dimensional or three-dimensional real-time anatomical image data of the patient from a KV-CBCT, MV-CT, or electronic field imaging device.

3. The real-time radiotherapy dose monitoring system according to claim 1, characterized in that: The data processing and fusion module includes: The parameter parsing unit is used to extract the actual position of the MLC blade and the actual angle of the frame from the machine operating parameters, and calculate their deviation from the planned values; The anatomical registration unit is used to match real-time images with planned CT images using a deformation image registration algorithm, and outputs a deformation field representing changes in organ position and density. The dose reconstruction unit is used to calculate the actual three-dimensional dose distribution based on the three-dimensional planned dose distribution, combined with the deviation and the deformation field, using a pre-calculated dose perturbation algorithm or a GPU-accelerated fast Monte Carlo algorithm.

4. The real-time radiotherapy dose monitoring system according to claim 3, characterized in that: The pre-calculated dose perturbation algorithm is as follows: a baseline dose cloud is pre-calculated based on the planned parameters using a high-precision algorithm; during treatment, the baseline dose cloud is linearly or nonlinearly perturbed according to real-time machine parameter deviations and anatomical changes to quickly output the corrected dose distribution.

5. The real-time radiotherapy dose monitoring system according to claim 1, characterized in that: The dose comparison and early warning module includes: A dose comparison unit is used to calculate the absolute dose difference and relative dose deviation between the three-dimensional actual dose distribution and the three-dimensional planned dose distribution; The gamma analysis unit is used to perform gamma analysis based on the pass rate index to assess the consistency of dose distribution in spatial and dose dimensions. The early warning unit is configured to activate the audiovisual alarm device and generate a suggestion to suspend treatment when the relative dose deviation exceeds 5% or the gamma pass rate is less than 95%.

6. The real-time radiotherapy dose monitoring system according to claim 1, characterized in that, Also includes: The data storage and report generation module is connected to the data interface module, data processing and fusion module, and dose comparison and early warning module. It is used to structurally store all real-time data, intermediate processing results, and final comparison results, and automatically generate structured reports that comply with clinical quality assurance standards.

Images