A data processing method, apparatus and equipment

Abstract

This invention provides a data processing method, apparatus, and device. The method includes: acquiring a VMAT plan; acquiring the corresponding CBCT projection and its header file of the patient during treatment of the patient by an accelerator based on the VMAT plan; calculating and determining the shape data formed by the MV source beam after being blocked by the MLC at each exit moment of the KV source based on the data in the VMAT plan and header file; at least inputting the patient's CBCT projection and the shape data into a target model to obtain a target projection image based on the target model, wherein the target projection image is an image without scattering contamination from the MV source beam and the KV source beam, and the target model is used to process the patient's CBCT projection and obtain the target projection image. The data processing method of this invention can quickly and efficiently process low-quality CBCT projections to obtain high-precision CBCT projection images of the patient.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a data processing method, device and storage medium. Background Technology

[0002] Cone-beam computed tomography (CBCT) is a widely used image-guided device, primarily used for pre-radiotherapy patient positioning and post-beam transmission verification imaging to quantify the effects of tumor movement and organ motion during fractionated treatment. However, CBCT images acquired before or after treatment do not record real-time anatomical information during beam transmission. To address this issue, volumetric modulated arc therapy (VMAT) has been widely adopted. Researchers began acquiring CBCT images of patients projected by kilovolt (kV) beams during beam transmission to obtain real-time three-dimensional anatomical information, enabling more accurate planning verification and in-treatment positioning correction. However, scattered photons from kV imaging sources lead to poor CBCT image quality. Furthermore, acquiring CBCT projections during VMAT treatment results in a large number of scattered photons from megavolt (MV) beams reaching the detector panel used for kV imaging, further degrading CBCT image quality and severely impacting the effectiveness of CBCT image acquisition during VMAT treatment. Therefore, correcting scattering contamination caused by kV and MV sources in CBCT images is crucial.

[0003] Currently, the following methods are used to improve the quality of CBCT images acquired during VMAT treatment:

[0004] 1. Add a physical blocker consisting of equally spaced lead wires and continuously move it between the projection source and the patient, estimate the scattered signals of KV and MV from the blocked area of ​​the detector panel, and insert it into the unblocked area.

[0005] 2. Acquire CBCT projection 1 with both KV and MV sources enabled, and CBCT projection 2 with only the MV source enabled. Subtract projection 2 from projection 1 to obtain the projection after removing MV source scattering contamination.

[0006] However, the above solutions have the following drawbacks:

[0007] Method 1 requires additional hardware, which is costly and not easy to promote. Method 2 requires two projections, which is impractical for actual clinical use. Furthermore, this method only removes the scattering effect caused by the MV source, but does not solve the scattering effect caused by the KV source, so the CBCT image quality is still not high. Summary of the Invention

[0008] This invention provides a data processing method, apparatus, and device that can rapidly and efficiently process low-quality CBCT projections to obtain high-precision CBCT projection images of patients.

[0009] To address the aforementioned technical problems, embodiments of the present invention provide a data processing method, the method comprising:

[0010] Obtain the volumetric rotating intensity-modulated virtual machine attack (VMAT) schedule;

[0011] Acquire the cone-beam computed tomography (CBCT) projection and its header file corresponding to the patient, acquired during the accelerator's treatment of the patient based on the VMAT plan;

[0012] Based on the data in the VMAT plan and header file, the shape data of the megavolt MV source beam after being blocked by the multi-leaf collimator MLC at each beam output time of the kV source is calculated and determined.

[0013] The patient's CBCT projection and shape data are input into a target model to obtain a target projection image based on the target model. The target projection image is an image free from MV source beam and KV source beam scattering contamination. The target model is used to process the patient's CBCT projection and obtain the target projection image.

[0014] As an optional embodiment, the acquisition of the volumetric rotating intensity-modulated arc device (VMAT) plan includes:

[0015] A VMAT plan for radiotherapy of the patient, prepared by a planning system, is obtained. The VMAT plan includes the gantry orientation, dose rate of the MV source beam, and distance between the center point of the accelerator and the MV source for each control point.

[0016] As an optional embodiment, the header file includes the shooting angle of the gantry when each frame of CBCT projection is acquired, the distance between the KV source and the CBCT detector panel and the center point, and the offset information of the CBCT detector panel relative to the center point under different shooting angles;

[0017] The calculation based on the VMAT plan and header file determines the shape data of the MV source beam after being blocked by the MLC at each beam exit moment of the KV source, including:

[0018] The beam angle of the KV source for each iteration is calculated and determined based on the data in the header file.

[0019] Based on the rack orientation recorded in the VMAT plan and the beam angle of the KV source for each time, determine the pair of control points closest to each beam angle;

[0020] With the motion rate of the MLC located between a pair of control points and the rate of change of the dose rate of the MV light source both fixed, interpolation is performed based on the information of each pair of control points recorded in the VMAT plan to obtain the shape data of the MV source beam after being blocked by the MLC at each beam exit moment of the KV source.

[0021] As an optional embodiment, the method further includes:

[0022] The three-dimensional structural data of the MV source beam after being blocked by the MLC are calculated based on the distance between the center point of each control point and the MV source, and the shape data corresponding to each beam exit time of the KV source.

[0023] Based on the three-dimensional structural data, the data in the header file, and the dose rate of the MV source beam corresponding to each shooting angle, a virtual CBCT projection corresponding to each shooting angle is calculated and determined. The virtual CBCT projection is a CBCT projection of a three-dimensional structure simulated by the three-dimensional structural data.

[0024] As an optional embodiment, the step of inputting at least the patient's CBCT projection and the shape data into the target model to obtain a target projection image based on the target model includes:

[0025] The shape data, the patient's CBCT projection, and the virtual CBCT projection are input into the target model to obtain the target projection image based on the target model.

[0026] As an optional embodiment, the target model is trained based on the following method:

[0027] The training data includes historical shape data of the MV source beam after being blocked by MLC at each exit moment of the KV source, determined based on historical data; CBCT projection of historical patients; historical virtual CBCT projection corresponding to the historical shape data; and historical target projection image containing the scanning information recorded in the CBCT projection of historical patients. The historical target projection image is an image without scattering contamination from the MV source beam and the KV source beam.

[0028] The target model is obtained by training a pre-defined model architecture based on the training data.

[0029] As an optional embodiment, it also includes:

[0030] Obtain CT images of the historical patients after the removal of the bed board;

[0031] Obtaining historical target projection images containing scan information recorded in the CBCT projections of historical patients includes:

[0032] The historical target projection image is calculated and determined based on the CT images, historical patients' CBCT projections, and data from historical header files.

[0033] As an optional embodiment, the step of calculating and determining the historical target projection image based on the CT image, the CBCT projection of historical patients, and the data in the historical header file includes:

[0034] CBCT images are generated by reconstructing images from the CBCT projections of historical patients.

[0035] Based on the CBCT image, the CT image is processed to obtain a deformed CT image, and the image content of the deformed CT image is matched with the image content structure of the CBCT image;

[0036] Based on the data in the historical header file, calculate the digital reconstructed ray projection obtained from the deformed CT image at each shooting angle, and form the historical target projection image based on the digital reconstructed ray projection.

[0037] The present invention also provides a data processing apparatus, comprising:

[0038] The first acquisition module is used to acquire the volumetric rotating intensity-modulated virtual attack (VMAT) plan.

[0039] The second acquisition module is used to acquire the cone-beam computed tomography (CBCT) projection of the patient and its header file, which are collected during the treatment of the patient by the accelerator based on the VMAT plan.

[0040] The calculation module is used to calculate and determine the shape data of the megavolt MV source beam after it is blocked by the multi-leaf collimator MLC at each beam output time of the kV source, based on the VMAT plan and the data in the header file.

[0041] The processing module is used to input at least the patient's CBCT projection and the shape data into the target model to obtain a target projection image based on the target model. The target projection image is an image without MV source beam and KV source beam scattering contamination. The target model is used to process the patient's CBCT projection and obtain the target projection image.

[0042] The present invention also provides an electronic device, including

[0043] One or more processors;

[0044] Memory, configured to store one or more programs;

[0045] When the one or more programs are executed by the one or more processors, the one or more processors implement the data processing method as described in any of the above embodiments.

[0046] Based on the disclosure of the above embodiments, it can be understood that the beneficial effects of the embodiments of the present invention include: by pre-constructing a target model for image processing, the device only needs to obtain the VMAT plan and the CBCT projection and its header file obtained from a rapid scan of the patient to calculate the input data of the target model. Then, by inputting the input data into the target model for processing, a high-precision CBCT projection image can be directly obtained, which can be directly used by doctors to diagnose diseases. The above process is not only fast and efficient, but also effectively removes the influence of MV source beams and KV source beams during the projection imaging process, ensuring that the obtained target projection image is free from photon scattering contamination, thereby maintaining high accuracy and precision. Furthermore, the above process does not add any new hardware; it is all implemented through software construction, resulting in low cost and strong feasibility. Attached Figure Description

[0047] Figure 1 This is a flowchart of the data processing method in an embodiment of the present invention.

[0048] Figure 2 This is a flowchart of a data processing method according to another embodiment of the present invention.

[0049] Figure 3 This is a flowchart of a data processing method according to another embodiment of the present invention.

[0050] Figure 4 This is a flowchart of a data processing method according to another embodiment of the present invention.

[0051] Figure 5 These are CBCT projection images in different states in embodiments of the present invention.

[0052] Figure 6 This is a structural block diagram of the data processing device in an embodiment of the present invention. Detailed Implementation

[0053] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings, but these are not intended to limit the scope of the invention.

[0054] It should be understood that various modifications can be made to the embodiments disclosed herein. Therefore, the following description should not be considered as limiting, but merely as an example of embodiments. Other modifications within the scope and spirit of this disclosure will be apparent to those skilled in the art.

[0055] The accompanying drawings, which are included in and form part of this specification, illustrate embodiments of the present disclosure and, together with the general description of the disclosure given above and the detailed description of the embodiments given below, serve to explain the principles of the disclosure.

[0056] These and other features of the invention will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.

[0057] It should also be understood that although the invention has been described with reference to some specific examples, those skilled in the art can certainly implement many other equivalent forms of the invention, which have the features described in the claims and are therefore all within the scope of protection defined herein.

[0058] The above and other aspects, features and advantages of this disclosure will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.

[0059] Specific embodiments of the present disclosure are described thereafter with reference to the accompanying drawings; however, it should be understood that the disclosed embodiments are merely examples of the present disclosure and can be implemented in various ways. Well-known and / or repeated functions and structures are not described in detail to avoid unnecessary or redundant details that could obscure the present disclosure. Therefore, the specific structural and functional details disclosed herein are not intended to be limiting, but merely to serve as the basis and representative basis for the claims to teach those skilled in the art to use the present disclosure in a variety of substantially any suitable detailed structures.

[0060] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in still another embodiment,” all of which may refer to one or more of the same or different embodiments according to this disclosure.

[0061] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0062] Figure 1 This is a flowchart of the data processing method in an embodiment of the present invention, such as... Figure 1 As shown, an embodiment of the present invention provides a data processing method, the method comprising:

[0063] S101: Obtain the volumetric rotating intensity-modulated virtual attack (VMAT) plan;

[0064] S102: Acquire the cone-beam computed tomography (CBCT) projection and its header file of the corresponding patient during the treatment of the patient based on the VMAT plan at the accelerator.

[0065] S103: Based on the data in the VMAT plan and header file, calculate and determine the shape data of the megavolt MV source beam after it is blocked by the multi-leaf collimator MLC at each beam output time of the kV source;

[0066] S104: At least the patient's CBCT projection and shape data are input into the target model to obtain a target projection image based on the target model. The target projection image is an image without MV source beam or KV source beam scattering contamination. The target model is used to process the patient's CBCT projection and obtain the target projection image.

[0067] For example, before a patient undergoes CBCT projection, the doctor will formulate an imaging and treatment plan based on the patient's actual condition, namely, a volumetric rotational intensity-modulated computed tomography (VMAT) plan. This plan records various operating parameters of the accelerator as a whole during the treatment process. Further, in this embodiment, the accelerator operates based on the VMAT plan. During the treatment of the patient, a cone-beam computed tomography (CBCT) projection of the patient is simultaneously acquired (this CBCT projection is a low-quality projection, an image interfered with by photon scattering, with low reference value). During this projection acquisition process, a header file corresponding to the CBCT projection is generated synchronously to record various parameters of the imaging equipment, projection light source, etc., during the acquisition process. After obtaining the above data, the equipment system calculates and determines the shape data of the megavolt (MV) source beam (based on the light source projected by the accelerator during patient treatment) after being blocked by the multi-leaf collimator (MLC) at each beam exit moment of the kilovolt (kV) source (projection light source during projection imaging). Under normal circumstances, the shape structure defined by this shape data should be the same as the lesion area and the area to be treated in the patient's body. Since the input data of the target model formed by the corresponding preprocessing has been obtained at this time, at least the CBCT projection and shape data can be input into the target model to form the target projection area based on the target model. The target projection area is an image that has been removed from the scattering and contamination of the MV source beam and KV source beam during the imaging and projection process. It can more accurately and clearly reflect the actual scanning information, so that the projection image has higher reference value and can be used by doctors for treatment.

[0068] Based on the disclosure of the above embodiments, it can be understood that the beneficial effects of this embodiment include: by pre-constructing a target model for image processing, the device only needs to obtain the VMAT plan and the CBCT projection and its header file obtained from a rapid scan of the patient to calculate the input data of the target model. Then, by inputting the input data into the target model for processing, a high-precision CBCT projection image can be directly obtained, which can be directly used by doctors to diagnose diseases. The above process is not only fast and efficient, but also effectively removes the influence of MV source beams and KV source beams during the projection imaging process, ensuring that the obtained target projection image is free from photon scattering contamination, thereby maintaining high accuracy and precision. Furthermore, the above process does not add any new hardware; it is all implemented through software construction, resulting in low cost and strong feasibility.

[0069] Specifically, in this embodiment, obtaining the volumetric rotating intensity-modulated arc therapy (VMAT) plan includes:

[0070] S201: Obtain the VMAT plan for radiotherapy of the patient prepared by the planning system. The VMAT plan includes the gantry orientation, dose rate of the MV source beam, and distance between the center point of the accelerator and the MV source for each control point.

[0071] The VMAT plan in this embodiment can be completed using the Pinnacle planning system, or other types of planning systems, depending on the specific requirements. Taking the Pinnacle planning system as an example, the Pinnacle planning system generates a text file for each formulated VMAT plan. Extracting this file yields the gantry angle of each control point in the VMAT plan, the dose rate of the MV source beam under different angles and other conditions, the shape of the multi-leaf collimator (MLC), and the distance from the center point of the accelerator to the MV source. This information can then be exported for use.

[0072] Furthermore, when the accelerator executes the VMAT plan, it can use the pre-set scanning mode of Elekta's XVI imaging system to acquire CBCT projections during MV source-based patient treatment. Moreover, after obtaining the target projection image, the XVI system can then reconstruct the CBCT image based on the target projection image and the header file of the original CBCT projection.

[0073] The header file in this embodiment includes the shooting angle of the gantry during each CBCT projection, the distances between the KV source and the CBCT detector panel and the center point, and the offset information of the CBCT detector panel relative to the center point at different shooting angles. The specific information recorded in the header file is not unique, but it includes at least the information recorded in this embodiment. For example, in addition to the above information, the header file may also record information such as whether the KV source is emitted at each shooting angle.

[0074] Among them, such as Figure 2 As shown, based on the VMAT plan and header file, the shape data of the MV source beam after being blocked by the MLC at each beam exit time of the KV source are calculated and determined, including:

[0075] S301: Calculate and determine the output angle of the KV source for each iteration based on the data in the header file;

[0076] S302: Based on the rack orientation and the output angle of the KV source for each time recorded in the VMAT plan, determine the pair of control points closest to each output angle;

[0077] S303: With the motion rate of the MLC located between a pair of control points and the rate of change of the dose rate of the MV source fixed, interpolation is performed based on the information of each pair of control points recorded in the VMAT plan to obtain the shape data of the MV source beam after being blocked by the multi-leaf collimator MLC at each beam exit moment of the KV source.

[0078] For example, since the accelerator is based on VMAT treatment, the entire accelerator equipment rotates around the patient. Therefore, the imaging gantry also rotates around the patient along with the accelerator. The projected images include multiple frames corresponding to various angles of the patient's 360°. The KV source emits a beam towards the patient at each imaging angle to complete the projection imaging. When calculating the shape data, the system first calculates and determines the actual beam angle of the KV source at each imaging based on the information recorded in the obtained header file. Then, based on the obtained VMAT plan, it determines the position of each recorded control point, corresponding to the accelerator gantry orientation of each control point (the VMAT plan file records relevant information about each control point). Afterward, based on the data determined above, the positions are determined in the same coordinate system to find the pair of control points closest to each KV source beam angle, i.e., the two control points. Once the nearest control point is determined, assuming the movement of the MLCs between control points is uniform and the rate of change of the dose rate of the MV source is also uniform, the equipment system can perform interpolation based on the information of each pair of control points recorded in the VMAT plan, including angles and dose rates, to obtain the shape data of the MV source beam after being blocked by the multi-leaf collimator MLC at each beam exit moment of the KV source, i.e., at the moment of capturing each frame of projection. This shape data can be represented as a planar image formed by multiplying the binarized two-dimensional data by the dose rate of the MV source.

[0079] Based on the shape data obtained above, the low-quality CBCT projection obtained from the patient's original imaging can be processed to filter out the projection image presented by the scattering effect of MV source and KV source beams on a large scale, and obtain a target projection image with almost no scattering effect of MV source and KV source.

[0080] However, to further improve the accuracy of the target projected image, optionally, such as Figure 3 As shown, the method in this embodiment further includes:

[0081] S401: Based on the distance between the center point of each control point and the MV source, and the shape data corresponding to each beam exit time of the KV source, calculate and determine the three-dimensional structure data of the MV source beam after it is blocked by the MLC.

[0082] S402: Based on the three-dimensional structural data, the data in the header file, and the dose rate of the MV source beam corresponding to each shooting angle, the virtual CBCT projection corresponding to each shooting angle is calculated and determined. The virtual CBCT projection is the CBCT projection of the three-dimensional structure simulated by the three-dimensional structural data.

[0083] In the above steps, the distance between the center point of each control point and the MV source can be determined based on the information recorded in the VMAT plan. Then, based on the shape data corresponding to each exit moment of the KV source calculated in the previous steps, the three-dimensional structure data of the MV source beam after being blocked by the MLC can be comprehensively calculated. The three-dimensional structure data forms a cone-like shape, with the tip pointing towards the MV source and the larger end facing the MLC. When this three-dimensional structure data is obtained, i.e., the three-dimensional structure is obtained, the geometric positional relationship between the KV source, the gantry, and the accelerator center point can be determined first based on the data in the header file. Then, combined with the three-dimensional structure, the three-dimensional structure is projected using the forward projection method to calculate and determine the candidate virtual CBCT projection for each shooting angle. After that, the candidate virtual CBCT projection is multiplied by the dose rate of the MV source beam corresponding to the same shooting angle to obtain the virtual CBCT projection that can correctly characterize the three-dimensional structure.

[0084] In this embodiment, when at least the patient's CBCT projection and shape data are input into the target model to obtain a target projection image based on the target model, it may include:

[0085] The shape data, the patient's CBCT projection, and the virtual CBCT projection are input into the target model to obtain the target projection image based on the target model.

[0086] As mentioned above, based on the shape data, most, or even all, of the influence of MV and KV sources can be removed. However, in order to ensure higher accuracy and stability of the projected image, this embodiment can further filter out MV and KV sources and retain the image on the original CBCT projection when processing the low-quality CBCT projection of the original patient, so as to obtain a CBCT projection that only retains the patient's scanning information, i.e., the target projection image.

[0087] Furthermore, the target model in this embodiment is a convolutional neural network model, such as a network structure using ResNet101 + SegNet to form the encoder and decoder, and selecting L1 distance as the loss function. Model training is performed based on the following method:

[0088] The training data includes historical shape data of the MV source beam after being blocked by MLC at each exit moment of the KV source, determined based on historical data; CBCT projection of historical patients; historical virtual CBCT projection of the corresponding historical shape data; and historical target projection image containing the scan information recorded in the CBCT projection of historical patients. The historical target projection image is an image without scattering contamination from the MV source beam and the KV source beam.

[0089] The target model is obtained by training a pre-defined model architecture based on the training data.

[0090] In other words, when preparing training data, the aforementioned morphological data, low-quality CBCT projections of patients, and virtual CBCT projections can be determined as input data for model training based on a large amount of historical data from different patients. The specific methods for obtaining these data are consistent with the steps described above and will not be repeated here. For the model's output data, i.e., uncontaminated, high-precision CBCT projections, these can be obtained using existing scanning methods that are too time-consuming, or they can be obtained using the following methods:

[0091] S501: Obtain CT images of historical patients after bed board removal;

[0092] Obtain historical target projection images containing scan information recorded in historical patients' CBCT projections, including:

[0093] S502: Calculate and determine the historical target projection image based on CT images, historical patient CBCT projections, and data from historical header files.

[0094] For example, after a patient undergoes a simulated CT scan, the resulting CT image is transmitted to the Pinnacle planning system or other processing software. This embodiment is based on the processing implemented using the planning system. In the planning system, the system outlines the human body in the CT image and then performs a slab removal process, which sets the CT values ​​of the parts outside the human body outline to zero and treats them as air. This yields the CT image after the slab removal process, which is then exported.

[0095] When calculating and determining historical target projection images based on CT images, historical patient CBCT projections, and data from historical header files, such as... Figure 4 As shown, it can be obtained using the following method:

[0096] S503: Reconstruct CBCT images based on historical patient CBCT projections;

[0097] S504: Based on CBCT image processing, CT images are obtained to obtain deformed CT images, and the image content of the deformed CT images is matched with the image content structure of the CBCT images;

[0098] S505: Calculates the digital reconstructed ray projection obtained from the deformed CT image at each shooting angle based on the data in the historical header file, and forms a historical target projection image based on the digital reconstructed ray projection.

[0099] For example, CBCT images are reconstructed based on historical patient CBCT projections to form CBCT images. Then, the CT image after removing the bed slab is deformed and registered onto the corresponding historical CBCT projection of a specific patient to obtain a deformed CT image. This ensures that the CT image and the CBCT image maintain as similar a structure as possible, i.e., image content structure matching. Subsequently, the geometrical positional information between the KV source, center point, and imaging equipment can be determined based on data in the historical header file to calculate the digital reconstructed ray projection (DRR) of the deformed CT at each imaging angle. Based on this DRR, the output of the target model is constructed. For example... Figure 5 As shown, image a represents a CT image, image b represents a low-quality CBCT projection image of the patient, and image c represents a target projection image formed after processing by the target model in this embodiment. By comparison, it can be clearly seen that the image processed by the target model of this application has significantly higher accuracy and clarity than images a and b, with outstanding effect.

[0100] Furthermore, to better illustrate the method of this embodiment, the following detailed description is provided in conjunction with practical applications:

[0101] Complete the patient's VMAT plan in the Pinnacle planning system and transfer the radiotherapy plan to the accelerator for execution. Export a text file generated by the Pinnacle planning system for each radiotherapy plan, recording the necessary information.

[0102] When the accelerator executes the radiotherapy plan, it selects the XVI pre-set scanning mode to acquire CBCT projections during MV beam therapy and export the CBCT projections and header files.

[0103] Export the above files and information or transmit them wirelessly to an external processing device, such as a server, for data processing.

[0104] 1) Based on the steps described above, the input data for the target model is obtained, specifically including the shape data, the patient's CBCT projection, and the virtual CBCT projection.

[0105] 2) After normalizing the input data, input it into the target model to quickly generate an ideal non-scattering projection image, i.e., the target projection image.

[0106] 3) Based on the data recorded in the header file, the generated target projection image is reconstructed using the FDK algorithm to obtain the descattered image.

[0107] 4) The image can be transmitted to an XVI system for positioning correction during treatment. It can also be transmitted to a treatment plan verification system to validate the effectiveness of the treatment plan. Physicians can also make diagnoses directly based on the image.

[0108] like Figure 6 As shown, another embodiment of the present invention also provides a data processing apparatus, including:

[0109] The first acquisition module is used to acquire the volumetric rotating intensity-modulated virtual attack (VMAT) plan.

[0110] The second acquisition module is used to acquire the cone-beam computed tomography (CBCT) projection of the patient and its header file, which are acquired during the treatment of the patient by the accelerator based on the VMAT plan.

[0111] The calculation module is used to calculate and determine the shape data of the megavolt MV source beam after it is blocked by the multi-leaf collimator MLC at each beam output time of the kV source, based on the VMAT plan and the data in the header file.

[0112] The processing module is used to input at least the patient's CBCT projection and the shape data into the target model to obtain a target projection image based on the target model. The target projection image is an image without MV source beam and KV source beam scattering contamination. The target model is used to process the patient's CBCT projection and obtain the target projection image.

[0113] As an optional embodiment, the method for obtaining a volumetric rotating intensity-modulated arc device (VMAT) plan includes:

[0114] A VMAT plan for radiotherapy of the patient is obtained by a planning system, the VMAT plan including the gantry orientation, dose rate of the MV source beam, and distance between the center point of the accelerator and the MV source for each control point.

[0115] As an optional embodiment, the header file includes the shooting angle of the gantry when each frame of CBCT projection is acquired, the distance between the KV source and the CBCT detector panel and the center point, and the offset information of the CBCT detector panel relative to the center point under different shooting angles;

[0116] The calculation based on the VMAT plan and header file determines the shape data of the MV source beam after being blocked by the MLC at each beam exit moment of the KV source, including:

[0117] The beam angle of the KV source for each iteration is calculated and determined based on the data in the header file.

[0118] Based on the rack orientation recorded in the VMAT plan and the beam angle of the KV source for each time, determine the pair of control points closest to each beam angle;

[0119] With the motion rate of the MLC located between a pair of control points and the rate of change of the dose rate of the MV light source both fixed, interpolation is performed based on the information of each pair of control points recorded in the VMAT plan to obtain the shape data of the MV source beam after being blocked by the multi-leaf collimator MLC at each beam exit moment of the KV source.

[0120] As an optional embodiment, the data processing apparatus further includes:

[0121] The first determining module is used to calculate and determine the three-dimensional structure data of the MV source beam after it is blocked by the MLC based on the distance between the center point of each control point and the MV source, and the shape data corresponding to each beam exit time of the KV source.

[0122] The second determining module is used to calculate and determine the virtual CBCT projection corresponding to each shooting angle based on the three-dimensional structure data, the data in the header file, and the dose rate of the MV source beam corresponding to each shooting angle. The virtual CBCT projection is a CBCT projection of a three-dimensional structure simulated by the three-dimensional structure data.

[0123] As an optional embodiment, the step of inputting at least the patient's CBCT projection and the shape data into the target model to obtain a target projection image based on the target model includes:

[0124] The shape data, the patient's CBCT projection, and the virtual CBCT projection are input into the target model to obtain the target projection image based on the target model.

[0125] As an optional embodiment, the target model is trained based on the following method:

[0126] The training data includes historical shape data of the MV source beam after being blocked by MLC at each exit moment of the KV source, determined based on historical data; CBCT projection of historical patients; historical virtual CBCT projection corresponding to the historical shape data; and historical target projection image containing the scanning information recorded in the CBCT projection of historical patients. The historical target projection image is an image without scattering contamination from the MV source beam and the KV source beam.

[0127] The target model is obtained by training a pre-defined model architecture based on the training data.

[0128] As an optional embodiment, the data processing apparatus further includes:

[0129] The third acquisition module is used to acquire CT images of the historical patient after the bed board was removed;

[0130] Obtaining a historical target projection image containing scan information recorded in the CBCT projection of a historical patient includes:

[0131] The historical target projection image is calculated and determined based on the CT images, historical patients' CBCT projections, and data from historical header files.

[0132] As an optional embodiment, the step of calculating and determining the historical target projection image based on the CT image, the CBCT projection of historical patients, and historical header files includes:

[0133] CBCT images are generated by reconstructing images based on the CBCT projections of historical patients.

[0134] Based on the CBCT image, the CT image is processed to obtain a deformed CT image, and the image content of the deformed CT image is matched with the image content structure of the CBCT image;

[0135] Based on the data in the historical header file, calculate the digital reconstructed ray projection obtained from the deformed CT image at each shooting angle, and form the historical target projection image based on the digital reconstructed ray projection.

[0136] Another embodiment of the present invention also provides an electronic device, comprising:

[0137] One or more processors;

[0138] Memory, configured to store one or more programs;

[0139] When the one or more programs are executed by the one or more processors, the one or more processors perform the above processing method.

[0140] An embodiment of the present invention also provides a storage medium storing a computer program thereon, which, when executed by a processor, implements the processing method described above. It should be understood that the various solutions in this embodiment have the corresponding technical effects in the above-described method embodiments, and will not be repeated here.

[0141] This invention also provides a computer program product, which is tangibly stored on a computer-readable medium and includes computer-readable instructions. When executed, these computer-executable instructions cause at least one processor to perform processing methods as described in the embodiments above. It should be understood that the various solutions in this embodiment have the corresponding technical effects in the above-described method embodiments, and will not be repeated here.

[0142] It should be noted that the computer storage medium of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. Computer-readable media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access storage media (RAM), read-only storage media (ROM), erasable programmable read-only storage media (EPROM or flash memory), optical fibers, portable compact disk read-only storage media (CD-ROM), optical storage media, magnetic storage media, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program configured for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, antenna, optical fiber, RF, etc., or any suitable combination thereof.

[0143] It should be understood that although this application is described according to various embodiments, not every embodiment contains only one independent technical solution. This way of describing the specification is only for clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

[0144] The above embodiments are merely exemplary embodiments of the present invention and are not intended to limit the present invention. The scope of protection of the present invention is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to the present invention within its spirit and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of the present invention.

Claims

1. A data processing method, characterized in that, The method includes: Obtain the volumetric rotating intensity-modulated virtual machine attack (VMAT) schedule; Acquire the cone-beam computed tomography (CBCT) projection and its header file corresponding to the patient, acquired during the accelerator's treatment of the patient based on the VMAT plan; Based on the data in the VMAT plan and header file, the shape data of the megavolt MV source beam after being blocked by the multi-leaf collimator MLC at each beam output moment of the kV source is calculated and determined. This shape data is a planar image formed by multiplying the binarized two-dimensional data by the dose rate of the MV source. At least the patient's CBCT projection and shape data are input into a target model to obtain a target projection image based on the target model. The target projection image is an image free from MV source beam and KV source beam scattering contamination. The target model is used to process the patient's CBCT projection and obtain the target projection image to remove the influence of MV source beam and KV source beam during the projection imaging process, so that the obtained target projection image is free from photon scattering contamination. The target model is a convolutional neural network model, trained based on the following method: The training data includes historical shape data of the MV source beam after being blocked by MLC at each exit moment of the KV source, determined based on historical data, historical CBCT projections of patients, and historical target projection images containing scan information recorded in the historical patients' CBCT projections. The historical target projection images are images without scattering contamination from the MV source beam and the KV source beam. The target model is obtained by training a pre-defined model architecture based on the training data.

2. The method according to claim 1, characterized in that, The acquisition of the volumetric rotating intensity-modulated virtual attack (VMAT) plan includes: A VMAT plan for radiotherapy of the patient, prepared by a planning system, is obtained. The VMAT plan includes the gantry orientation, dose rate of the MV source beam, and distance between the center point of the accelerator and the MV source for each control point.

3. The method according to claim 2, characterized in that, The header file includes the shooting angle of the gantry when each frame of CBCT projection is acquired, the distance between the KV source and the CBCT detector panel and the center point, and the offset information of the CBCT detector panel relative to the center point under different shooting angles; The calculation based on the VMAT plan and header file determines the shape data of the MV source beam after being blocked by the MLC at each beam exit moment of the KV source, including: The beam angle of the KV source for each iteration is calculated and determined based on the data in the header file. Based on the rack orientation recorded in the VMAT plan and the beam angle of the KV source for each time, determine the pair of control points closest to each beam angle; With the motion rate of the MLC located between a pair of control points and the rate of change of the dose rate of the MV source both fixed, interpolation is performed based on the information of each pair of control points recorded in the VMAT plan to obtain the shape data of the MV source beam after being blocked by the MLC at each beam exit moment of the KV source.

4. The method according to claim 3, characterized in that, The method further includes: The three-dimensional structural data of the MV source beam after being blocked by the MLC are calculated based on the distance between the center point of each control point and the MV source, and the shape data corresponding to each beam exit time of the KV source. Based on the three-dimensional structural data, the data in the header file, and the dose rate of the MV source beam corresponding to each shooting angle, a virtual CBCT projection corresponding to each shooting angle is calculated and determined. The virtual CBCT projection is a CBCT projection of a three-dimensional structure simulated by the three-dimensional structural data.

5. The method according to claim 4, characterized in that, The step of inputting at least the patient's CBCT projection and shape data into the target model to obtain a target projection image based on the target model includes: The shape data, the patient's CBCT projection, and the virtual CBCT projection are input into the target model to obtain the target projection image based on the target model.

6. The method according to claim 4, characterized in that, The training data also includes historical virtual CBCT projections corresponding to the historical shape data.

7. The method according to claim 6, characterized in that, Also includes: Obtain CT images of the historical patients after the removal of the bed board; Obtaining historical target projection images containing scan information recorded in the CBCT projections of historical patients includes: The historical target projection image is calculated and determined based on the CT images, historical patients' CBCT projections, and data from historical header files.

8. The method according to claim 7, characterized in that, The step of calculating and determining the historical target projection image based on the CT image, historical patients' CBCT projections, and data from historical header files includes: CBCT images are generated by reconstructing images based on the CBCT projections of historical patients. Based on the CBCT image, the CT image is processed to obtain a deformed CT image, and the image content of the deformed CT image is matched with the image content structure of the CBCT image; Based on the data in the historical header file, the digital reconstructed ray projection obtained from the deformed CT image at each shooting angle is calculated, and the historical target projection image is formed based on the digital reconstructed ray projection.

9. A data processing apparatus, characterized in that, include: The first acquisition module is used to acquire the volumetric rotating intensity modulated arc therapy (VMAT) plan. The second acquisition module is used to acquire the cone-beam computed tomography (CBCT) projection of the patient and its header file, which are collected during the process of the accelerator treating the patient based on the VMAT plan. The calculation module is used to calculate and determine the shape data of the megavolt MV source beam after it is blocked by the multi-leaf collimator MLC at each beam output time of the kV source, based on the data in the VMAT plan and header file. The shape data is a planar image formed by multiplying the binarized two-dimensional data by the dose rate of the MV source. A processing module is used to input at least the patient's CBCT projection and shape data into a target model to obtain a target projection image based on the target model. The target projection image is an image free from MV source beam and KV source beam scattering contamination. The target model is used to process the patient's CBCT projection and obtain the target projection image to remove the influence of MV source beam and KV source beam during the projection imaging process, so that the obtained target projection image is free from photon scattering contamination. The target model is a convolutional neural network model, trained based on the following method: The training data includes historical shape data of the MV source beam after being blocked by MLC at each exit moment of the KV source, determined based on historical data, historical CBCT projections of patients, and historical target projection images containing scan information recorded in the historical patients' CBCT projections. The historical target projection images are images without scattering contamination from the MV source beam and the KV source beam. The target model is obtained by training a pre-defined model architecture based on the training data.

10. An electronic device, characterized in that, include One or more processors; Memory, configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the data processing method as described in any one of claims 1-8.

Images