Radiotherapy bolus dose feedback and tracking system based on image registration

Abstract

The invention relates to the field of navigation signal processing, and discloses a radiation therapy bolus dose feedback and tracking system based on image registration, the system comprises an image acquisition module, an image registration module, a bolus identification module, a dose recalculation module, a dose deviation analysis module and a feedback and adjustment module, space registration is carried out through an FBCT image acquired every day and an original plan CT, and the radiation therapy bolus dose feedback and tracking system based on image registration is obtained. The fitting state between the bolus and the skin is automatically identified, the position of an air gap is detected, and dose recalculation is carried out based on an actual structure. Through comparison with an original planned dose, the system can output a deviation index of a key dose parameter and provide visual feedback, if the deviation exceeds a set threshold value, a prompt is automatically sent out, and a radiotherapy technician and a physician are supported to carry out bolus adjustment or call an adaptive planning engine to reconstruct a treatment plan. Quantitative analysis of bolus fitting quality and closed-loop control of dose delivery consistency are realized, radiotherapy precision and safety are effectively improved, and the method is suitable for superficial radiotherapy scenes containing bolus design.

Description

Technical Field

[0001] The present invention relates to the technical field of navigation signal processing, and more particularly to a radiotherapy bolus dose feedback and tracking system based on image registration. Background Art

[0002] Radiation therapy is a key treatment modality for breast cancer, often requiring the use of boluses (sensitizers) to increase the radiation dose to the skin and superficial tissues. However, the quality of bolus placement during each treatment session is easily affected by multiple factors, including patient position, operator experience, and the bolus's own shape. This can often lead to air gaps or misalignment between the bolus and the skin. This suboptimal fit can severely disrupt dose distribution, resulting in insufficient dose to the target area, potentially reducing treatment efficacy and increasing the risk of local recurrence, while also increasing toxicity and side effects in normal tissues.

[0003] Although modern radiotherapy equipment (such as the United Imaging uRT-linac506c) is equipped with daily diagnostic-grade FBCT imaging, enabling the monitoring of intraday anatomical changes, clinical practice still lacks effective automated tools to accurately assess the actual placement of boluses and their specific impact on dose. Current methods rely heavily on subjective visual inspection, which is inefficient, difficult to quantify, and unable to quickly determine the extent of actual dose deviations. Furthermore, there is a lack of a mechanism for systematically tracking and analyzing these daily deviations.

[0004] Therefore, existing technologies have obvious deficiencies in addressing bolus placement variability and are unable to form closed-loop feedback to optimize the treatment process. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a radiotherapy bolus dose feedback and tracking system based on image registration to solve the problems existing in the above-mentioned background technology.

[0006] The present invention provides the following technical solution: a radiotherapy bolus dose feedback and tracking system based on image registration, comprising:

[0007] Image acquisition module: used to obtain FBCT images of patients in bolus state before or during each radiotherapy;

[0008] Image registration module: It is used to spatially register the acquired daily FBCT images with the original planned CT images used when formulating the patient's radiotherapy plan, establish the spatial correspondence between the daily anatomical structure and the planned anatomical structure, and output the registered daily FBCT images;

[0009] Bolus recognition module: This module extracts image features from the registered daily FBCT images and analyzes the extracted image features to identify key areas related to the bolus and the patient's anatomical structure information for that day.

[0010] Dose recalculation module: Based on the identified anatomical structure information of the patient on the daily FBCT image, the dose is recalculated for the patient's anatomical structure on the daily basis, and the actual dose distribution under the anatomical structure on the daily basis is recalculated;

[0011] Dose deviation analysis module: compares and analyzes the actual dose distribution obtained by dose recalculation with the originally planned dose distribution, and calculates the dose deviation value between the actual dose and the originally planned dose;

[0012] Feedback and adjustment module: generates feedback information based on the dose deviation analysis results.

[0013] Preferably, the image acquisition module places the patient on the treatment bed in accordance with the requirements of the treatment plan and places the bolus material on the designated area of ​​the patient's skin. While the patient maintains the treatment position and the bolus is placed, the FBCT scanning program integrated in the radiotherapy equipment is started to obtain the patient's three-dimensional cone-beam CT image on the day, and the obtained three-dimensional cone-beam CT image of the patient on the day is directly uploaded to the computer terminal through a dedicated network protocol.

[0014] Preferably, the image registration module verifies and standardizes the format of the collected FBCT images, processes the bolus materials in the images, and then performs hierarchical registration of the patient's daily FBCT images with the original planned CT images used when formulating the radiotherapy plan, including coarse registration, fine registration, and fine-tuning registration. The registration data in the image registration process is analyzed for real-time evaluation of the registration quality in the registration process, and the anatomical structure of the planned CT is mapped to the FBCT through the deformation field. Finally, the registered FBCT image is output, which contains the spatial information of the planned anatomical structure. The planned dose distribution is then mapped to the current anatomical structure through the deformation field, and the actual delivered dose is calculated and compared with the planned dose for real-time monitoring of the registered FBCT image and dose distribution. If the dose deviation exceeds the clinical threshold, real-time correction of the FBCT image is triggered.

[0015] Preferably, the bolus recognition module first extracts features from the FBCT image through image processing technology, analyzes the extracted image features in combination with the CT value features in the image, identifies key areas related to the bolus, and then uses a threshold segmentation method to outline the key areas related to the bolus. Within the outlined key areas related to the bolus, the anatomical structure information of the patient on that day is further identified, wherein the anatomical structure information of the patient on that day includes the bolus area, the outer contour of the patient's skin, and the potential air gap between the bolus and the skin.

[0016] Preferably, the dose recalculation module aligns the daily FBCT image with the original treatment plan CT through a rigid or deformable registration algorithm, assigns contrast agent density according to the automatic segmentation result, detects and assigns air density, maps grayscale values ​​to density values ​​through the HU value-electron density calibration curve, and generates a three-dimensional density map of the anatomical structure of the day based on the density assignment as input for dose calculation. Based on the density map of the anatomical structure of the day generated by the registered FBCT image, the beam parameters set in the original radiotherapy plan are used to recalculate the actual dose distribution under the anatomical structure of the day on the density map corresponding to the anatomical structure of the day, so as to evaluate the dose distribution under the actual geometric state of the day.

[0017] Preferably, the dose deviation analysis module extracts dose parameters from the recalculated dose distribution of the day, including the coverage dose of the target area, the maximum dose, and the maximum dose or specific volume dose of the organ at risk, and compares the recalculated actual dose parameters of the day with the original planned dose parameters to calculate the dose deviation value, determine whether the treatment of the day meets the clinical goals, set the deviation threshold according to the clinical goals and treatment plan, and trigger manual intervention in the treatment process if the dose deviation value exceeds the deviation threshold; if the dose deviation value is less than or equal to the deviation threshold, continue to perform the current treatment operation.

[0018] Preferably, the feedback information of the feedback and adjustment module at least includes displaying daily dose deviation and triggering subsequent adjustment processes, automatically generating a feedback report based on the dose deviation analysis results, and displaying a daily deviation map in a visual manner to indicate the specific location where the deviation occurs, and generating a trend graph of dose deviation changes with treatment fractions, outputting adjustment suggestions on bolus placement or fit, and automatically triggering an adaptive plan reconstruction process when the deviation is serious or significant deviation occurs in multiple consecutive fractions.

[0019] Technical effects and advantages of the present invention:

[0020] (1) Based on the spatial registration of the collected daily FBCT images with the original planned CT images used when formulating the patient's radiotherapy plan, and the image feature extraction and analysis of the registered daily FBCT images, the key areas related to the bolus and the anatomical structure information of the patient on that day were identified, which is conducive to the automatic and quantitative evaluation of the actual placement effect of the bolus (including air gap) and its impact on dose distribution through daily three-dimensional images (FBCT).

[0021] (2) Based on the anatomical structure information of the patient on the day of the identified daily FBCT images, the dose of the patient's anatomical structure information on the day is recalculated, and the actual dose distribution obtained by the dose recalculation is compared and analyzed with the originally planned dose distribution, and the dose deviation value between the actual dose and the originally planned dose is calculated. This is conducive to a more accurate understanding of the actual dose received by the patient, and can detect potential dose deviations caused by bolus problems at an early stage, avoiding the significant impact of deviation accumulation on the treatment effect.

[0022] (3) Based on the dose deviation analysis results, feedback information is generated, providing specific feedback information and adjustment suggestions, which helps therapists optimize daily bolus placement operations and provides objective dosimetric basis for whether more complex interventions are needed, further improving the safety and effectiveness of treatment, thereby improving clinical work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a system structure diagram of the present invention.

[0024] Figure 2 1 is a diagram of the method steps of this embodiment.

[0025] Figure 3 FIG. 4 is a schematic diagram of the image registration process of FBCT and planning CT in this embodiment.

[0026] Figure 4 This is a flow chart of the trend feedback and prompt mechanism of this embodiment. DETAILED DESCRIPTION

[0027] The technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the present invention. In addition, the forms of the various structures described in the following embodiments are merely examples. The image registration-based radiotherapy bolus dose feedback and tracking system involved in the present invention is not limited to the various structures described in the following embodiments. All other embodiments obtained by ordinary technicians in this field without making creative work fall within the scope of protection of the present invention.

[0028] like Figure 1This embodiment provides a radiotherapy bolus dose feedback and tracking system based on image registration, including:

[0029] Image acquisition module: used to obtain FBCT images of patients in bolus state before or during each radiotherapy.

[0030] In this embodiment, the image acquisition module positions the patient on the treatment bed and places the bolus material on the designated area of ​​the patient's skin in accordance with the requirements of the treatment plan. While the patient maintains the treatment position and the bolus is placed, the FBCT scanning program integrated into the radiotherapy equipment is activated to obtain the patient's three-dimensional cone-beam CT image for the day. The obtained three-dimensional cone-beam CT image of the patient for the day is directly uploaded to the computer terminal via a dedicated network protocol.

[0031] It should be specifically noted that, according to the radiotherapy plan, the patient is placed on the treatment bed and ensured to be in the correct treatment position, and a bolus material is placed on the designated area of ​​the patient's skin. While the patient maintains the treatment position and the bolus material is properly placed, the cone-beam CT device is started to acquire an FBCT image of the patient's current treatment position. The FBCT device rotates the X-ray source and detector to perform a 360-degree scan around the patient to acquire a three-dimensional volumetric image. The acquired FBCT image can reflect the patient's current actual anatomical structure, including the position and shape of the bolus and its relationship with surrounding tissues.

[0032] Image registration module: It is used to spatially register the acquired daily FBCT images with the original planned CT images used when formulating the patient's radiotherapy plan, establish the spatial correspondence between the daily anatomical structure and the planned anatomical structure, and output the registered daily FBCT images.

[0033] In this embodiment, the image registration module verifies and standardizes the format of the collected FBCT images, processes the bolus materials in the images, and then performs hierarchical registration of the patient's daily FBCT images with the original planned CT images used when formulating the radiotherapy plan, including coarse registration, fine registration, and fine-tuning registration. The registration data in the image registration process is analyzed for real-time evaluation of the registration quality in the registration process, and the anatomical structure of the planned CT is mapped to the FBCT through the deformation field. Finally, the registered FBCT image containing the spatial information of the planned anatomical structure is output, and the planned dose distribution is mapped to the current anatomical structure through the deformation field. The actual delivered dose is calculated and compared with the planned dose for real-time monitoring of the registered FBCT image and dose distribution. If the dose deviation exceeds the clinical threshold, real-time correction of the FBCT image is triggered.

[0034] It should be specifically explained that by verifying the metadata of the FBCT images uploaded to the computer (such as patient ID, scanning date, and body position mark), checking the image quality, converting the FBCT images and the original planned CT images into a standardized format, adjusting the window width of the FBCT images and the planned CT images, and matching the grayscale distribution of the anatomical structures (such as bones and soft tissues), firstly coarsely aligning the daily FBCT images, marking the bolus area in the FBCT image by the threshold segmentation method, generating a bolus mask, and performing density calibration on the bolus area, thereby eliminating To eliminate the dose calculation error caused by the change of material thickness or density, a rigid body transformation (such as affine transformation) is selected as the coarse registration algorithm to spatially align the patient's daily FBCT images with the original planned CT images used in the formulation of the radiotherapy plan. The specific operation process of the coarse registration includes: automatically detecting the bone area as the registration seed point, aligning the bone structure of the FBCT and the planning CT by translation + rotation to quickly eliminate the global error, calculating the bone overlap (mutual information) and global error, evaluating the quality of the coarse registration, and displaying the bone overlap and global error of the registered image. If the mutual information 0.8 or global error 5mm, trigger manual intervention to adjust the registration parameters; if local anatomical structure differences still exist after coarse registration, continue with fine registration. The specific operation process includes: based on coarse registration, the Demons algorithm based on mutual information is used to perform non-rigid registration on the CT image containing the bolus, and a deformation field generation algorithm (such as B-spline interpolation) is used to generate a deformation field to compensate for local anatomical structure differences. The deformation field is applied to the CT image and the bolus mask to ensure accurate alignment of the bolus area with the surrounding anatomical structures; if the target boundary mismatch or key structure alignment error is still large after fine registration, fine-tuning registration is triggered, and feature point-based registration or semi-automatic registration is used to optimize the registration accuracy of the target boundary. Indicators such as the target Dice coefficient and Hausdorff distance are calculated to evaluate the anatomical structure matching accuracy after fine-tuning registration. If the matching accuracy meets the standard, the registration is completed; otherwise, the registration parameters or methods need to be readjusted to establish the spatial correspondence between the two images to achieve accurate matching of the anatomical structures;

[0035] The anatomical structure of the planned CT scan (such as the target volume and organs at risk) is mapped to the FBCT image space through the deformation field generated by coarse and fine registration. The registered FBCT image containing the spatial information of the planned anatomical structure is output. The overlap index is then used to verify whether the mapped anatomical structure matches the FBCT anatomical structure. If not, the registration parameters are adjusted and the deformation field is regenerated. If they match, the dose distribution of the mapped anatomical structure is mapped and the dose deviation is calculated.

[0036] The planned dose distribution is mapped to the current anatomical structure through the deformation field by converting the coordinates of each voxel into the FBCT space and interpolating the new dose value. The mapped dose distribution is used to reflect the dose transfer under the actual anatomical structure. The dose deviation is calculated by comparing the mapped dose distribution with the planned dose distribution. If the dose deviation exceeds the clinical threshold (such as the target area D95 deviation >5%, the dose limit of the organ at risk is exceeded), the FBCT image is corrected in real time based on the target area offset and Bolus deformation to compensate for the anatomical structure changes, and the dose distribution is recalculated until the dose deviation meets the clinical requirements to ensure the accuracy of the dose distribution under the actual anatomical structure.

[0037] Bolus recognition module: Based on the image feature extraction of the registered daily FBCT images, the extracted image features are analyzed to identify the key areas related to the bolus and the anatomical structure information of the patient on that day.

[0038] In this embodiment, the bolus recognition module first extracts features from the FBCT image through image processing technology, analyzes the extracted image features in combination with the CT value features in the image, identifies key areas related to the bolus, and then uses a threshold segmentation method to outline the key areas related to the bolus. Within the outlined key areas related to the bolus, the anatomical structure information of the patient on that day is further identified. The anatomical structure information of the patient on that day includes the bolus area, the outer contour of the patient's skin, and the potential air gap between the bolus and the skin.

[0039] Specifically, FBCT images were standardized using image processing techniques, including scanning parameter correction, motion artifact removal, and grayscale normalization. HU value deviations caused by different scanning devices were corrected using a linear regression model. Respiratory motion was compensated using the PROST algorithm based on projection data. FBCT images were grayscale normalized to extract texture features. CT value features were directly extracted using the original HU values. Texture features were extracted from grayscale-normalized images. Geometric features were extracted from candidate regions using region growing or watershed algorithms. Image features were jointly analyzed with CT value features in the image. A hierarchical decision rule was constructed to identify key areas associated with the bolus. A global threshold was calculated based on the Otsu algorithm. Combined with the typical HU range of the bolus, key areas associated with the bolus were delineated.

[0040] Import the CT image data and parse it into an image grayscale matrix, denoted as ; The value of each element in corresponds to the grayscale value (such as HU value) of the corresponding pixel in the image.

[0041] The method for identifying and analyzing the patient's skin contour is to set the minimum grayscale threshold of the body. , for the image matrix Each element in Make a judgment, if , then in the body mask matrix The corresponding position Assigned to 1; otherwise assigned to 0, that is: ,if , ,if , The outer boundary of the skin (which can be extracted by edge detection algorithm or morphological gradient operation) is defined as the outer contour of the skin;

[0042] The identification and analysis method for the bolus area is: set the lower limit of the grayscale value range of the bolus and upper limit (This range is determined by the specific bolus material properties), the image matrix Each element in Make a judgment; if , then in the bolus mask matrix The corresponding position Assigned to 1; otherwise assigned to 0, that is: ,if , , for other situations;

[0043] The method for identifying and analyzing the potential air gap between the bolus and the skin is to set the maximum grayscale threshold of the air gap. , identifying air gap pixels The following conditions must usually be met: (1) The pixel grayscale value is lower than the threshold: ; (2) Pixels are located in the body mask In addition: ; (3) The pixel is spatially adjacent to the skin contour and bolus area (This can be done through morphological operations, such as and After the expansion operation, find the area that meets conditions 1 and 2, or calculate the pixel to and The pixels that meet all the above conditions are In the air gap mask matrix The corresponding position is assigned a value of 1, and the rest of the positions are assigned a value of 0, and the significant air gap area between the bolus and the skin is automatically detected and outlined.

[0044] Dose recalculation module: Based on the identified anatomical structure information of the patient on the daily FBCT image, the dose is recalculated for the anatomical structure information of the patient on the daily basis, and the actual dose distribution under the anatomical structure on the daily basis is recalculated.

[0045] In this embodiment, the dose recalculation module aligns the daily FBCT image with the original treatment plan CT through a rigid or deformable registration algorithm, assigns contrast agent density according to the automatic segmentation result, detects and assigns air density, maps grayscale values ​​to density values ​​through the HU value-electron density calibration curve, and generates a three-dimensional density map of the anatomical structure of the day based on the density assignment as input for dose calculation. Based on the density map of the anatomical structure of the day generated by the registered FBCT image, the beam parameters set in the original radiotherapy plan are used to recalculate the actual dose distribution under the anatomical structure of the day on the density map corresponding to the anatomical structure of the day, so as to evaluate the dose distribution under the actual geometric state of the day.

[0046] It should be specifically explained that by integrating the registered FBCT image information, a density map representing the actual situation of the day is generated. In this density map, the identified bolus area is assigned the density corresponding to its material, the air gap area is assigned the air density, and the patient tissue is converted into electron density based on the grayscale value of FBCT to construct the anatomical model of the day; by loading all relevant beam parameters in the original radiotherapy plan, including beam energy, gantry angle, collimator angle, MLC leaf position sequence, each sub-field weight (MonitorUnits, MU), etc.; by using the same dose calculation algorithm as that used to formulate the original plan (such as Monte Carlo, convolution superposition, etc.), on the generated anatomical model of the day, the beam irradiation process of the original plan is simulated, and the actual dose distribution under the anatomical structure of the day is calculated. The specific calculation formula is: ,in, Indicates location The dosage, represents the radiation flux with energy E, Indicates location The energy at is the linear attenuation coefficient of E, Indicates location The tissue density, represents the reference density, and the linear attenuation coefficient It is related to the electron density of the tissue and the radiation energy, which can be obtained through the HU value-electron density calibration curve. From the 3D density map, the radiation flux Determined by the beam parameters in the original treatment plan.

[0047] Dose deviation analysis module: compares and analyzes the actual dose distribution obtained by dose recalculation with the originally planned dose distribution, and calculates the dose deviation value between the actual dose and the originally planned dose.

[0048] In this embodiment, the dose deviation analysis module extracts dose parameters from the recalculated dose distribution of the day, including the coverage dose of the target area, the maximum dose, and the maximum dose or specific volume dose of the organ at risk, and compares the recalculated actual dose parameters of the day with the original planned dose parameters to calculate the dose deviation value, determine whether the treatment of the day meets the clinical objectives, and set the deviation threshold according to the clinical objectives and treatment plan. If the dose deviation value exceeds the deviation threshold, manual intervention in the treatment process is triggered. If the dose deviation value is less than or equal to the deviation threshold, the current treatment operation continues.

[0049] It should be noted that the dose volume histogram (DVH) data and key dosimetric parameters (such as the volume percentage of the target area receiving a specific dose) of the predefined target area (such as PTV, CTV) and organs at risk (OAR) are automatically extracted from the original planned dose distribution and the recalculated dose distribution on the day. , the average dose of the target area , maximum dose to target area , the volume percentage of the organ at risk receiving a specific dose The deviation between the actual dose and the planned dose of these key parameters on the day is calculated as , generate a dose difference map, and intuitively display the specific spatial location and degree of increase or decrease of the daily dose relative to the planned dose on the patient image by color coding and other means.

[0050] Feedback and adjustment module: generates feedback information based on the dose deviation analysis results.

[0051] In this embodiment, the feedback information of the feedback and adjustment module at least includes displaying daily dose deviations and triggering subsequent adjustment processes, automatically generating a feedback report based on the dose deviation analysis results, and visually displaying a daily deviation map to indicate the specific location where the deviation occurs. It also generates a trend graph of dose deviations changing with treatment fractions, outputs adjustment suggestions for bolus placement or fit, and automatically triggers an adaptive plan reconstruction process when the deviation is severe or significant deviations occur in multiple consecutive fractions.

[0052] It should be specifically noted that based on the results of the dose deviation analysis, easy-to-understand feedback information is generated. This information should at least include daily deviation values ​​for key dosimetric parameters. Furthermore, a visual dose deviation map (showing the spatial location of the deviation) and a trend graph showing the accumulation or change of dose deviation with treatment fractions can be provided. In addition, based on the size and location of the deviation, the system can output adjustment suggestions for bolus placement. When the detected dose deviation exceeds the preset safety threshold, or a significant and consistent deviation trend appears across multiple consecutive treatment fractions, the module can automatically mark the patient / fraction, or directly trigger an alarm, prompting the clinician or physicist to intervene, or even initiate the adaptive radiation therapy (ART) process, which is to re-optimize or formulate the treatment plan based on the current cumulative dose or significant single-shot deviation.

[0053] like Figure 2 This embodiment provides a radiotherapy bolus dose feedback and tracking method based on image registration, including:

[0054] Step 1: Obtain FBCT images of the patient in the bolus state before or during each radiotherapy treatment;

[0055] Step 2: Perform spatial registration of the acquired daily FBCT images with the original planned CT images used when formulating the patient's radiotherapy plan, establish a spatial correspondence between the daily anatomical structures and the planned anatomical structures, and output the registered daily FBCT images;

[0056] Step 3: Extract image features from the registered daily FBCT images and analyze the extracted image features to identify key areas related to the bolus and the patient's anatomical structure information on that day;

[0057] Step 4: Based on the identified anatomical structure information of the patient on the daily FBCT image, the dose is recalculated for the anatomical structure information of the patient on the daily basis, and the actual dose distribution under the anatomical structure on the daily basis is recalculated;

[0058] Step 5: Compare and analyze the actual dose distribution obtained by dose recalculation with the originally planned dose distribution, and calculate the dose deviation between the actual dose and the originally planned dose;

[0059] Step 6: Generate feedback information based on the dose deviation analysis results.

[0060] like Figure 3The figure shows the image registration process of FBCT and planning CT in this embodiment. The specific process includes: starting the registration module and selecting the registration method to perform rigid registration (bone landmarks) and elastic registration (soft tissue + Bolus deformation) operations, loading the image, executing the registration algorithm to perform rigid registration and elastic registration, performing translation and rotation operations through rigid registration, adjusting B-spline / Demons / DIR, etc. through elastic registration, and real-time evaluation of the registration quality of the rigid registration and elastic registration processes to determine whether the registration is satisfactory. If so, the registered FBCT image and deformation field (DeformationField) are directly output. If not, the parameters are adjusted and re-registered.

[0061] like Figure 4 The figure shows the trend feedback and prompt mechanism process of this embodiment. The specific process includes: when the daily dose deviation analysis is completed, it is determined whether the deviation exceeds the safety threshold. If the deviation exceeds the safety threshold, visual feedback information is generated, an alarm is triggered / abnormal fractionation is marked, and then it is determined whether Bolus adjustment is required. If Bolus adjustment is required, the Bolus adjustment suggestion is executed, and the FBCT image is reacquired and the dose is calculated. If Bolus adjustment is not required, it is further determined whether the ART trigger condition is met. If the trigger condition is met, a single severe deviation or a continuous trend is analyzed, and the adaptive radiotherapy ART process is started. The plan is re-optimized based on the current data, the treatment parameters are updated and the new plan is executed. Finally, conventional treatment is continued. If the trigger condition is not met, conventional treatment is directly performed. If the deviation does not exceed the safety threshold, conventional treatment is continued.

[0062] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0063] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. Radiotherapy bolus dose feedback and tracking system based on image registration, characterized by: include: Image acquisition module: used to obtain FBCT images of patients in bolus state before or during each radiotherapy; Image registration module: It is used to spatially register the acquired daily FBCT images with the original planned CT images used when formulating the patient's radiotherapy plan, establish the spatial correspondence between the daily anatomical structure and the planned anatomical structure, and output the registered daily FBCT images; Bolus recognition module: This module extracts image features from the registered daily FBCT images and analyzes the extracted image features to identify key areas related to the bolus and the patient's anatomical structure information for that day. Dose recalculation module: Based on the identified anatomical structure information of the patient on the daily FBCT image, the dose is recalculated for the patient's anatomical structure on the daily basis, and the actual dose distribution under the anatomical structure on the daily basis is recalculated; Dose deviation analysis module: compares and analyzes the actual dose distribution obtained by dose recalculation with the originally planned dose distribution, and calculates the dose deviation value between the actual dose and the originally planned dose; Feedback and adjustment module: generates feedback information based on the dose deviation analysis results.

2. The radiotherapy bolus dose feedback and tracking system based on image registration according to claim 1, characterized in that: The image acquisition module places the patient on the treatment bed in accordance with the requirements of the treatment plan and places the bolus material on the designated area of ​​the patient's skin. While the patient maintains the treatment position and the bolus is placed, the FBCT scanning program integrated in the radiotherapy equipment is started to obtain the patient's three-dimensional cone-beam CT image on the day, and the obtained three-dimensional cone-beam CT image of the patient on the day is directly uploaded to the computer terminal via a dedicated network protocol.

3. The radiotherapy bolus dose feedback and tracking system based on image registration according to claim 2, characterized in that: The image registration module verifies and standardizes the format of the collected FBCT images, processes the bolus materials in the images, and then performs hierarchical registration of the patient's daily FBCT images with the original planned CT images used when formulating the radiotherapy plan, including coarse registration, fine registration, and fine-tuning registration. The registration data in the image registration process is analyzed for real-time evaluation of the registration quality in the registration process, and the anatomical structure of the planned CT is mapped to the FBCT through the deformation field. Finally, the registered FBCT image containing the spatial information of the planned anatomical structure is output, and the planned dose distribution is mapped to the current anatomical structure through the deformation field. The actual delivered dose is calculated and compared with the planned dose for real-time monitoring of the registered FBCT image and dose distribution. If the dose deviation exceeds the clinical threshold, real-time correction of the FBCT image is triggered.

4. The radiotherapy bolus dose feedback and tracking system based on image registration according to claim 3, characterized in that: The bolus recognition module first extracts features from the FBCT image using image processing technology, identifies key areas related to the bolus based on an analysis of the extracted image features combined with the CT value features in the image, and then uses a threshold segmentation method to outline the key areas related to the bolus. Within the outlined key areas related to the bolus, the anatomical structure information of the patient on that day is further identified. The anatomical structure information of the patient on that day includes the bolus area, the outer contour of the patient's skin, and potential air gaps between the bolus and the skin.

5. The radiotherapy bolus dose feedback and tracking system based on image registration according to claim 4, characterized in that: The dose recalculation module aligns the daily FBCT image with the original treatment plan CT through a rigid or deformable registration algorithm, assigns contrast agent density according to the automatic segmentation result, detects and assigns air density, maps grayscale values ​​to density values ​​through the HU value-electron density calibration curve, and generates a three-dimensional density map of the anatomical structure of the day based on the density assignment as input for dose calculation. Based on the density map of the anatomical structure of the day generated by the registered FBCT image, the beam parameters set in the original radiotherapy plan are used to recalculate the actual dose distribution under the anatomical structure of the day on the density map corresponding to the anatomical structure of the day, so as to evaluate the dose distribution under the actual geometric state of the day.

6. The radiotherapy bolus dose feedback and tracking system based on image registration according to claim 5, characterized in that: The dose deviation analysis module extracts dose parameters from the recalculated dose distribution of the day, including the coverage dose of the target area, the maximum dose, and the maximum dose or specific volume dose of the organ at risk. The recalculated actual dose parameters of the day are compared with the original planned dose parameters to calculate the dose deviation value, determine whether the treatment of the day meets the clinical goals, and set the deviation threshold according to the clinical goals and treatment plan. If the dose deviation value exceeds the deviation threshold, manual intervention in the treatment process is triggered. If the dose deviation value is less than or equal to the deviation threshold, the current treatment operation continues.

7. The radiotherapy bolus dose feedback and tracking system based on image registration according to claim 6, characterized in that: The feedback information of the feedback and adjustment module at least includes displaying daily dose deviations and triggering subsequent adjustment processes, automatically generating a feedback report based on the dose deviation analysis results, and visually displaying a daily deviation map to indicate the specific location where the deviation occurred. It also generates a trend graph of dose deviations changing with treatment fractions, outputs adjustment suggestions for bolus placement or fit, and automatically triggers an adaptive plan reconstruction process when the deviation is severe or significant deviations occur in multiple consecutive fractions.

Images