Dynamic tumor tracking method and apparatus, computer device, and storage medium

Abstract

The present application relates to a dynamic tumor tracking method and apparatus, a computer device, and a storage medium. The method comprises: acquiring 4DCT image data of a patient, and performing mixed data augmentation on the 4DCT image data to generate 3DCT images representing various respiratory phases of the lungs; on the basis of the 3DCT images, using a mathematical simulation method to generate simulated X-ray images corresponding to a respiratory cycle of the patient; registering the X-ray images with the 4DCT image data to generate a tumor tracking model of the patient; and inputting X-ray images of the patient during radiotherapy into the tumor tracking model, and outputting tumor tracking results by means of the tumor tracking model. The present application enables real-time three-dimensional tumor tracking by using a single low-dose X-ray image under arbitrary geometric conditions, can provide high-precision tumor localization on the premise of ensuring a low radiation dose, and is applicable to arbitrary projection angles and varying source-object distance conditions, allowing flexible adaptation to various geometric relationship changes in a treatment process.

Description

A dynamic tumor tracking method, device, computer equipment, and storage medium Technical Field

[0001] This application belongs to the field of medical image processing technology, and specifically relates to a dynamic tumor tracking method, device, computer equipment, and storage medium. Background Technology

[0002] During radiotherapy, the target area of ​​cancer patients changes due to physiological factors such as respiratory and cardiac movements. To ensure accurate delivery of the treatment dose to the target area and avoid damage to the patient's normal tissues, dynamic tumor tracking is necessary to monitor the target area's position in real time. Traditional techniques, such as CBCT (Cone Beam Computed Tomography), used for tumor tracking, suffer from time delays and high radiation doses, making it difficult to respond in real time to changes in tumor position caused by physiological movements like respiration, thus limiting its widespread clinical application. While the application of deep learning in medical image processing has significantly improved the accuracy of tumor tracking, existing deep learning models typically assume fixed geometry and cannot handle changes in the distance between the source and the object, or the projection angle. In actual clinical applications, especially in dynamic image-guided radiotherapy, the distance between the source and the object fluctuates due to changes in patient position, respiratory movements, or other physiological changes, making it difficult for existing deep learning models to adapt to these geometric variations.

[0003] Currently, mainstream tumor tracking methods in radiotherapy mainly include tumor tracking guided by CBCT and CT (Computed Tomography) images, and tumor tracking based on 2D / 3D image registration. CBCT and CT-based tumor tracking relies on acquiring multi-angle cone-beam projections during radiotherapy to reconstruct 3D images, which are then registered with the planning CT to locate the tumor. However, CBCT imaging requires acquiring hundreds of projection images, significantly increasing the patient's radiation dose. Furthermore, the imaging process is time-consuming and cannot meet the needs of real-time dynamic monitoring, especially when the target area changes position due to respiration or heartbeat. Tumor tracking based on 2D / 3D image registration utilizes 2D beam projections from radiotherapy with planning CT images to achieve tumor tracking, effectively reducing radiation dose and shortening processing time. To achieve matching between 2D images and the 3D target area, linear dimensionality reduction methods such as PCA (Principal Component Analysis) are commonly used. However, PCA struggles to capture complex nonlinear motions, thus limiting registration accuracy when dealing with irregular physiological movements such as respiration and heartbeat, making it difficult to meet clinical requirements. Summary of the Invention

[0004] This application provides a dynamic tumor tracking method, apparatus, computer device, and storage medium, which aims to at least partially solve one of the aforementioned technical problems in the prior art.

[0005] To address the above problems, this application provides the following technical solution:

[0006] A dynamic tumor tracking method, comprising:

[0007] Acquire the patient's 4DCT image data and perform hybrid data enhancement on the 4DCT image data to generate 3DCT images representing various respiratory stages of the lungs.

[0008] Based on the 3DCT images, X-ray images simulating the patient's respiratory cycle are generated using mathematical simulation methods.

[0009] The X-ray and 4DCT image data are registered to generate a tumor tracking model for the patient.

[0010] The X-ray images of the patient during radiotherapy are input into the tumor tracking model, and the tumor tracking results are output through the tumor tracking model.

[0011] The technical solution adopted in this application embodiment further includes: after acquiring the patient's 4DCT image data, it also includes:

[0012] Any one phase of the 4DCT image data is selected as a floating image, and the remaining N phases are used as reference images. The reference images and the floating images are then registered using an image registration method to obtain N deformation fields.

[0013] The technical solution adopted in this application embodiment further includes: performing hybrid data enhancement on the 4DCT image data to generate 3DCT images representing various respiratory stages of the lungs, specifically:

[0014] Two deformation fields are randomly selected from the N deformation fields and superimposed to obtain the interphase deformation.

[0015] Thin-plate spline interpolation is used to simulate random deformation in any phase;

[0016] The phase deformation and random deformation are superimposed with arbitrary weights to obtain a deformation field. The deformation field is then applied to the floating image to obtain 3D CT images representing various respiratory stages of the lungs.

[0017] The technical solution adopted in this application embodiment also includes: generating X-ray images simulating the patient's respiratory cycle based on the 3DCT images using a mathematical simulation method, specifically:

[0018] The 3D CT images were simulated using the Monte Carlo simulation method to generate 2D DRR, which was then used to simulate X-ray images during the patient's respiratory cycle.

[0019] The technical solution adopted in this application embodiment also includes: registering the X-ray image and 4DCT image data to generate a tumor tracking model for the patient, specifically as follows:

[0020] The X-ray image is upscaled to a three-dimensional feature map using residual blocks, and the three-dimensional feature map is encoded and parsed from different angles using a 3D encoding and decoding network. The three-dimensional feature map and the floating image are then registered.

[0021] The technical solution adopted in this application embodiment also includes: the tumor tracking model adopts a deep learning model that combines a 3D encoding / decoding network and the Unet-KAN mechanism.

[0022] The technical solution adopted in this application embodiment further includes: inputting the X-ray images of the patient during radiotherapy into the tumor tracking model, and outputting the tumor tracking results through the tumor tracking model, specifically as follows:

[0023] A single X-ray image from any angle is acquired during the patient's radiotherapy process. The single X-ray image is then input into the patient's tumor tracking model. The three-dimensional location of the tumor and organs at risk is obtained through the tumor tracking model, and the tumor tracking results are output.

[0024] Another technical solution adopted in this application embodiment is: a dynamic tumor tracking device, comprising:

[0025] Image acquisition module: used to acquire the patient's 4DCT image data and perform hybrid data enhancement on the 4DCT image data to generate 3DCT images representing various respiratory stages of the lungs;

[0026] Image simulation module: used to generate simulated X-ray images of the patient's respiratory cycle based on the 3DCT images using mathematical simulation methods;

[0027] Image registration module: used to register the X-ray images and 4DCT image data to generate a tumor tracking model for the patient;

[0028] Tumor tracking module: This module is used to input X-ray images of the patient during radiotherapy into the tumor tracking model and output tumor tracking results through the tumor tracking model.

[0029] Another technical solution adopted in this application embodiment is: a computer device, the computer device including a processor and a memory coupled to the processor, wherein,

[0030] The memory stores program instructions for implementing the dynamic tumor tracking method;

[0031] The processor is used to execute the program instructions stored in the memory to control the dynamic tumor tracking method.

[0032] Another technical solution adopted in this application embodiment is: a storage medium storing processor-executable program instructions, the program instructions being used to execute the dynamic tumor tracking method.

[0033] Compared to existing technologies, the beneficial effects of the embodiments of this application are as follows: The dynamic tumor tracking method, device, computer equipment, and storage medium of the embodiments of this application combine deep learning with 2D-3D registration technology, and use the patient's 4DCT image data to train a tumor tracking model specifically for that patient. A single ultra-low-dose X-ray image from any angle during the patient's radiotherapy is input into the tumor tracking model, achieving efficient real-time tumor tracking. The tumor tracking model of the embodiments of this application can achieve real-time three-dimensional tracking of the tumor under any geometric conditions using a single low-dose X-ray image. It can provide high-precision tumor localization while ensuring low radiation dose, and is applicable to any projection angle and source-object distance variation conditions. It can flexibly cope with various geometric relationship changes during treatment and adapt to more clinical application scenarios. The embodiments of this application can achieve more precise radiotherapy dose control under low-dose X-ray images, improving treatment efficacy while reducing damage to surrounding normal tissues. Attached Figure Description

[0034] Figure 1 is a flowchart of a dynamic tumor tracking method according to an embodiment of this application;

[0035] Figure 2 is a schematic diagram of a tumor tracking model according to an embodiment of this application;

[0036] Figure 3 is a schematic diagram of the tumor registration results from the end of expiration to the end of inspiration for lung cancer patient data;

[0037] Figure 4 is a schematic diagram of the tumor centroid motion error at different angles;

[0038] Figure 5 is a schematic diagram of the tumor registration results of the digital phantom XCAT from the end of expiration to the end of inspiration;

[0039] Figure 6 is a schematic diagram of the trajectory of the tumor centroid in the digital phantom;

[0040] Figure 7 is a schematic diagram of the structure of the dynamic tumor tracking device according to an embodiment of this application;

[0041] Figure 8 is a schematic diagram of the computer device structure according to an embodiment of this application;

[0042] Figure 9 is a schematic diagram of the structure of the storage medium according to an embodiment of this application. Detailed Implementation

[0043] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0044] The terms "first," "second," and "third" in this application are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. All directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of this application are only used to explain the relative positional relationships and movements between components in a specific orientation (as shown in the figures). If the specific orientation changes, the directional indications also change accordingly. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or computer device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or computer devices.

[0045] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0046] Specifically, please refer to Figure 1, which is a flowchart of the dynamic tumor tracking method according to an embodiment of this application. The dynamic tumor tracking method according to an embodiment of this application includes the following steps:

[0047] S100: Acquires the patient's 4DCT image data and performs hybrid data enhancement processing on the 4DCT image data to generate 3DCT images representing various respiratory stages of the lungs.

[0048] In this step, 4DCT refers to a low-dose imaging technique that sequentially records respiratory motion trajectories on CT images. It records lung CT images over multiple respiratory cycles using rapid imaging technology, and then combines these images with respiratory motion signals to form a three-dimensional image sequence of a portion of the respiratory cycle, creating a four-dimensional dataset. This allows for dynamic observation of the location, shape, and size of the tumor during respiratory motion.

[0049] Furthermore, 4DCT image data typically includes multiple phases, and the specific steps for hybrid data enhancement processing of 4DCT image data include:

[0050] S101: Select any phase of the 4DCT image data as the floating image, and use the remaining N phases as the reference image;

[0051] S102: Use image registration methods to register the reference image and the floating image to obtain N deformation fields;

[0052] S103: Randomly select two deformation fields from N deformation fields and superimpose them to obtain the inter-phase deformation, and use thin plate spline interpolation to simulate the random deformation of any phase;

[0053] S104: A large number of deformation fields are obtained by superimposing phase deformation and random deformation with arbitrary weights. The large number of deformation fields are applied to the floating image to obtain a large number of 3D CT images representing various respiratory stages of the lungs.

[0054] It is understood that the embodiments of this application generate a large number of 3D CT images covering all expiratory phases of the lungs by performing data augmentation processing on 4D CT image data, thereby providing rich training data for the model and improving the model's adaptability to respiratory movements.

[0055] S110: Based on 3DCT images, a mathematical simulation method is used to generate 2D DRR (Digitally Reconstructed Radiograph), which simulates X-ray images at different stages of the patient's respiratory cycle.

[0056] In this step, DRR generates simulated X-ray images from 3D CT images through mathematical simulation. This not only generates a large number of 2D images from multiple angles and distances, but also simulates different noise levels and artifact characteristics, which helps improve the model's generalization ability in clinical settings and provides the model with data that is closer to actual clinical situations. The mathematical simulation methods include, but are not limited to, Monte Carlo simulation algorithms (a mathematical technique based on random numbers).

[0057] S120: Performs image registration on X-ray images and floating images to generate a tumor tracking model for the patient;

[0058] In this step, the tumor tracking model employs a deep learning model combining a 3D encoding / decoding network and the Unet-KAN mechanism, as shown in Figure 2. The registration method between the X-ray image and the floating image is as follows: first, the X-ray image is upscaled to a three-dimensional feature map using residual blocks; then, the 3D encoding / decoding network encodes and parses the three-dimensional feature map from different angles; finally, the three-dimensional feature map and the floating image are registered, mapping the X-ray image to the three-dimensional location of the tumor and organs at risk in the three-dimensional image. The combination of the 3D encoding / decoding network and the Unet-KAN mechanism in the tumor tracking model enhances the ability to capture spatial information, thereby improving the accuracy of tumor tracking. Furthermore, this embodiment encodes the three-dimensional feature map from different angles using a 3D encoding / decoding network, enabling the tumor tracking model to cope with changes in source-target distance and angle caused by physiological changes such as patient position, respiration, or heart function, ensuring tracking accuracy under different geometric changes and enhancing the generalization performance of the tumor tracking model.

[0059] Based on the above, in this embodiment of the application, before radiotherapy, a tumor tracking model specifically for the patient is trained using hybrid data augmentation, Monte Carlo simulation, and two-dimensional / three-dimensional registration algorithms on the patient's 4DCT image data. This enables the tumor tracking model to accurately map the two-dimensional ultra-low dose X-ray projection image during radiotherapy to the three-dimensional target area and the location of organs at risk. The tumor tracking model can accurately generate the conversion relationship from two-dimensional image to three-dimensional location, achieving efficient and real-time tumor tracking and ensuring high-precision positioning based on the patient's individual information.

[0060] S130: Acquire a single X-ray image from any angle during the patient's radiotherapy process, input the single X-ray image into the patient's tumor tracking model, and output the three-dimensional position of the tumor and organs at risk through the tumor tracking model;

[0061] In this step, a single ultra-low-dose X-ray image from any angle during the patient's radiotherapy is acquired and input into the patient's tumor tracking model. The tumor tracking model then rapidly analyzes the X-ray image, outputting the three-dimensional location of the tumor and organs at risk in real time. This embodiment of the application achieves efficient real-time tumor tracking by obtaining the target area location in a short time using only a single low-dose X-ray image. Simultaneously, it significantly reduces the patient's radiation dose and minimizes damage to surrounding normal tissues, overcoming the clinical limitations of traditional C-image-guided methods due to high dose and delay.

[0062] To verify the feasibility and effectiveness of the embodiments of this application, experiments were conducted on a public dataset of lung cancer patients and on the digital phantom XCAT. Figure 3 shows a schematic diagram of tumor registration results from the end of expiration to the end of inspiration in lung cancer patient data. Odd-numbered columns represent unregistered images, and even-numbered rows represent registered images. It can be seen that both the lungs and tumors achieved good registration results. The average accuracy of DSC for the lungs was 0.97, while the overall tumor localization time was approximately 100 milliseconds, indicating that the embodiments of this application can achieve real-time low-dose 3D tumor tracking. Figure 4 shows a schematic diagram of tumor centroid motion error at different angles. It can be seen that the low-dose X-ray error at most angles is within 1 mm, achieving good tumor tracking results. Figure 5 shows a schematic diagram of tumor registration results from the end of expiration to the end of inspiration in the digital phantom XCAT, and Figure 6 shows a schematic diagram of the tumor centroid motion trajectory in the digital phantom. It can be seen that the embodiments of this application can also achieve good tumor tracking on the digital phantom.

[0063] Based on the above, the dynamic tumor tracking method of this application combines deep learning with 2D-3D registration technology. It trains a tumor tracking model specifically for the patient using the patient's 4DCT image data. A single ultra-low-dose X-ray image from any angle during radiotherapy is input into the tumor tracking model, achieving efficient real-time tumor tracking. The tumor tracking model of this application can achieve real-time three-dimensional tumor tracking using a single low-dose X-ray image under any geometric conditions. It provides high-precision tumor localization while ensuring low radiation dose and is applicable to any projection angle and source-object distance variations. It can flexibly handle various geometric changes during treatment and adapt to more clinical application scenarios. This application embodiment can achieve more precise radiotherapy dose control under low-dose X-ray imaging, improving treatment efficacy while reducing damage to surrounding normal tissues.

[0064] Please refer to Figure 7, which is a schematic diagram of the structure of the dynamic tumor tracking device according to an embodiment of this application. The dynamic tumor tracking method device 40 of this embodiment includes:

[0065] Image acquisition module 41: used to acquire the patient's 4DCT image data and perform hybrid data enhancement on the 4DCT image data to generate 3DCT images representing various respiratory stages of the lungs;

[0066] Image simulation module 42: used to generate simulated X-ray images of a patient's respiratory cycle based on the 3DCT images using mathematical simulation methods;

[0067] Image registration module 43: used to register the X-ray image and 4DCT image data to generate a tumor tracking model for the patient;

[0068] Tumor tracking module 44: used to input X-ray images of the patient during radiotherapy into the tumor tracking model, and output tumor tracking results through the tumor tracking model.

[0069] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0070] The apparatus provided in this application can be applied to the foregoing method embodiments. For details, please refer to the description of the above method embodiments, which will not be repeated here.

[0071] Please refer to Figure 8, which is a schematic diagram of the structure of a computer device according to an embodiment of this application. The computer device 50 includes:

[0072] Memory 51 storing executable program instructions;

[0073] Processor 52 connected to memory 51;

[0074] The processor 52 is used to call the executable program instructions stored in the memory 51 and perform the following steps: acquire the patient's 4DCT image data, and perform hybrid data enhancement on the 4DCT image data to generate 3DCT images representing various respiratory stages of the lungs; based on the 3DCT images, generate X-ray images simulating the patient's respiratory cycle using a mathematical simulation method; register the X-ray images and 4DCT image data to generate a tumor tracking model for the patient; input the X-ray images during the patient's radiotherapy process into the tumor tracking model, and output the tumor tracking results through the tumor tracking model.

[0075] The processor 52 can also be referred to as a CPU (Central Processing Unit). The processor 52 may be an integrated circuit chip with signal processing capabilities. The processor 52 can also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. A general-purpose processor can be a microprocessor or any conventional processor.

[0076] Please refer to Figure 9, which is a schematic diagram of the structure of the storage medium according to an embodiment of this application. The storage medium of this embodiment stores program instructions 61 capable of implementing the following steps: acquiring 4DCT image data of a patient, performing hybrid data enhancement on the 4DCT image data to generate 3DCT images representing various respiratory stages of the lungs; generating X-ray images simulating the patient's respiratory cycle based on the 3DCT images using a mathematical simulation method; registering the X-ray images and 4DCT image data to generate a tumor tracking model for the patient; inputting the X-ray images during the patient's radiotherapy process into the tumor tracking model, and outputting tumor tracking results through the tumor tracking model. The program instructions 61 can be stored in the aforementioned storage medium in the form of a software product, including several instructions to cause a computer device (which may be a personal computer, server, or network computer device, etc.) or processor to execute all or part of the steps of the methods of various embodiments of this application. The aforementioned storage media include: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program instructions, or terminal computer devices such as computers, servers, mobile phones, and tablets. Servers can be standalone servers or cloud servers providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0077] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, apparatuses, or units, and may be electrical, mechanical, or other forms.

[0078] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units. The above are merely embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A dynamic tumor tracking method, characterized in that, include: Acquire the patient's 4DCT image data and perform hybrid data enhancement on the 4DCT image data to generate 3DCT images representing various respiratory stages of the lungs. Based on the 3DCT images, mathematical simulation methods were used to generate X-ray images simulating the patient's respiratory cycle. The X-ray and 4DCT image data are registered to generate a tumor tracking model for the patient. The X-ray images of the patient during radiotherapy are input into the tumor tracking model, and the tumor tracking results are output through the tumor tracking model.

2. The dynamic tumor tracking method according to claim 1, characterized in that, After acquiring the patient's 4DCT image data, the process also includes: Any one phase of the 4DCT image data is selected as a floating image, and the remaining N phases are used as reference images. The reference images and the floating images are then registered using an image registration method to obtain N deformation fields.

3. The dynamic tumor tracking method according to claim 2, characterized in that, The process of performing hybrid data enhancement on the 4DCT image data to generate 3DCT images representing various respiratory stages of the lungs specifically involves: Two deformation fields are randomly selected from the N deformation fields and superimposed to obtain the interphase deformation. Thin-plate spline interpolation is used to simulate random deformation in any phase; The inter-phase deformation and random deformation are superimposed with arbitrary weights to obtain a deformation field. The deformation field is then applied to the floating image to obtain 3D CT images representing various respiratory stages of the lungs.

4. The dynamic tumor tracking method according to any one of claims 1 to 3, characterized in that, The process of generating simulated X-ray images during the patient's respiratory cycle based on the 3DCT images using mathematical simulation methods is as follows: The 3D CT images were simulated using the Monte Carlo simulation method to generate 2D DRR, which was then used to simulate X-ray images during the patient's respiratory cycle.

5. The dynamic tumor tracking method according to claim 4, characterized in that, The process of registering the X-ray and 4DCT image data to generate a tumor tracking model for the patient specifically involves: The X-ray image is upscaled to a three-dimensional feature map using residual blocks, and the three-dimensional feature map is encoded and parsed from different angles using a 3D encoding and decoding network. The three-dimensional feature map and the floating image are then registered.

6. The dynamic tumor tracking method according to claim 5, characterized in that, The tumor tracking model employs a deep learning model that combines a 3D encoder-decoder network and the Unet-KAN mechanism.

7. The dynamic tumor tracking method according to claim 6, characterized in that, The process of inputting X-ray images of the patient during radiotherapy into the tumor tracking model and outputting tumor tracking results through the tumor tracking model specifically involves: A single X-ray image from any angle is acquired during the patient's radiotherapy process. The single X-ray image is then input into the patient's tumor tracking model. The three-dimensional location of the tumor and organs at risk is obtained through the tumor tracking model, and the tumor tracking results are output.

8. A dynamic tumor tracking device, characterized in that, include: Image acquisition module: used to acquire the patient's 4DCT image data and perform hybrid data enhancement on the 4DCT image data to generate 3DCT images representing various respiratory stages of the lungs; Image simulation module: used to generate simulated X-ray images of the patient's respiratory cycle based on the 3DCT images using mathematical simulation methods; Image registration module: used to register the X-ray images and 4DCT image data to generate a tumor tracking model for the patient; Tumor tracking module: This module is used to input X-ray images of the patient during radiotherapy into the tumor tracking model and output tumor tracking results through the tumor tracking model.

9. A computer device, characterized in that, The computer device includes a processor and a memory coupled to the processor, wherein, The memory stores program instructions for implementing the dynamic tumor tracking method according to any one of claims 1-7; The processor is used to execute the program instructions stored in the memory to control the dynamic tumor tracking method.

10. A storage medium, characterized in that, The device stores processor-executable program instructions for performing the dynamic tumor tracking method according to any one of claims 1 to 7.

Images