Real-time image guidance method, apparatus, and system, radiotherapy system

Abstract

The application provides a real-time image guidance method, device and system, and a radiotherapy system, and belongs to the technical field of radiotherapy. The image guidance system can perform reliable real-time image guidance on a target object based on a target reference image and a real-time projection image of the target object. Since the target reference image is acquired after positioning is completed, the accuracy of real-time image guidance on the target object based on the target reference image and the real-time projection image is high.

Description

Technical Field

[0001] This disclosure relates to the field of radiotherapy technology, and in particular to a real-time image-guided method, device and system, and radiotherapy system. Background Technology

[0002] In radiotherapy settings, image-guided radiation therapy (IGRT) technology can be used to locate and track the position of the target object (e.g., a patient's tumor) in real time to guide the radiotherapy.

[0003] In related technologies, the IGRT system used to implement IGRT technology can acquire two-dimensional projection images of the target object at two different angles during treatment. These two-dimensional projection images can then be registered with reference images at the corresponding angles to determine two two-dimensional offsets. Finally, the three-dimensional offset of the target object can be calculated based on these two determined two-dimensional offsets, enabling the tracking of the target object's position. The reference image can be an image reconstructed by the IGRT system from the computed tomography (CT) images of the target object when generating the treatment plan.

[0004] However, the image-guided methods of related technologies have low accuracy and poor flexibility in tracking position. Summary of the Invention

[0005] This disclosure provides a real-time image-guided method, apparatus, system, and radiotherapy system, which can solve the problems of low tracking accuracy and poor flexibility in related technologies. The technical solution is as follows:

[0006] On the one hand, a real-time image guidance method is provided, the method comprising:

[0007] Acquire a target reference image of the target object, wherein the target reference image is an image determined based on the image acquired by the image guidance device after the target object has been positioned;

[0008] The image guidance device is used to acquire a real-time projected image of the target object;

[0009] Based on the target reference image and the real-time projected image, the target object is guided by real-time image.

[0010] On the other hand, a real-time image guidance device is provided, the device comprising:

[0011] The first acquisition module is used to acquire a target reference image of the target object, wherein the target reference image is an image determined based on the image acquired by the image guidance device after the target object has been positioned.

[0012] The second acquisition module is used to acquire a real-time projected image of the target object using the image guidance device;

[0013] The image guidance module is used to provide real-time image guidance to the target object based on the target reference image and the real-time projected image.

[0014] In another aspect, a real-time image guidance system is provided, the real-time image guidance system comprising: an image guidance device, a processor, and a memory;

[0015] The image guidance device is used to acquire images, and the memory stores instructions that are loaded and executed by the processor to implement the real-time image guidance method as described above.

[0016] In another aspect, a storage medium is provided that stores instructions that, when the storage medium is run on a processing component, cause the processing component to execute the real-time image guidance method as described above.

[0017] In another aspect, a radiotherapy system is provided, the radiotherapy system comprising: a patient support device, a main unit, and a real-time image guidance system; the real-time image guidance system is the system described above.

[0018] The host is connected to both the real-time image guidance system and the patient support device. The real-time image guidance system sends the target offset of the determined target object to the host, and the host adjusts the position of the patient support device based on the target offset.

[0019] The technical solutions provided in this disclosure have at least the following beneficial effects:

[0020] In summary, the embodiments of this disclosure provide a real-time image-guided method, apparatus, system, and radiotherapy system. The image-guided system can reliably guide a target object in real-time based on an acquired target reference image and a real-time projected image. Since the target reference image is acquired after positioning, the accuracy of real-time image guidance based on this target reference image and the real-time projected image is high.

[0021] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of the structure of a radiotherapy system provided in an embodiment of this disclosure;

[0024] Figure 2 This is a flowchart of a real-time image guidance method provided in an embodiment of this disclosure;

[0025] Figure 3 This is a flowchart of another real-time image guidance method provided in this embodiment of the disclosure;

[0026] Figure 4 This is a flowchart of a method for determining a target reference image provided in an embodiment of this disclosure;

[0027] Figure 5 This is a schematic diagram of a marker provided in an embodiment of this disclosure;

[0028] Figure 6 This is a flowchart of an image guidance method provided in an embodiment of this disclosure;

[0029] Figure 7 This is a flowchart of another image guidance method provided in this embodiment;

[0030] Figure 8 This is a flowchart of another real-time image guidance method provided in this embodiment of the disclosure;

[0031] Figure 9 This is a flowchart of yet another real-time image guidance method provided in this disclosure embodiment;

[0032] Figure 10 This is a block diagram of a real-time image guidance device provided in an embodiment of this disclosure;

[0033] Figure 11 This is a block diagram of another real-time image guidance device provided in the embodiments of this disclosure;

[0034] Figure 12 This is a block diagram of a real-time image guidance system provided in an embodiment of this disclosure.

[0035] The accompanying drawings have illustrated specific embodiments of this disclosure, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this disclosure to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of this disclosure clearer, the embodiments of this disclosure will be described in further detail below with reference to the accompanying drawings.

[0037] Figure 1 This is a schematic diagram of the structure of a radiotherapy system provided in an embodiment of this disclosure. Figure 1 As shown, the radiotherapy system may include a patient support device 01, a host 02, a real-time image guidance system 03, and a radiotherapy device 04.

[0038] Optionally, the patient support device 01 can be Figure 1 The treatment bed shown can be a 3D treatment bed, a 6D treatment bed, or other devices for supporting the patient, such as a treatment chair. The host unit 02 can be a control device. The image guidance system 03 can be an IGRT system. The host unit 02 can establish communication connections with both the patient support device 01 and the image guidance system 03. Figure 1 The connection shown can be wired or wireless. During radiotherapy, the image guidance system 03 can use IGRT technology to track the position of the target object (e.g., the patient's tumor) in real time and send the offset of the target object to the host 02. The host 02 can flexibly adjust the position of the patient support device 01 based on the received offset to achieve reliable real-time image guidance for the patient.

[0039] Optionally, the real-time image guidance system 03 may include an image guidance device 031, which may be a cone beam computed tomography (CBCT) device. That is, the image guidance device 031 may be used to acquire CBCT images of the target object.

[0040] For example, refer to Figure 1 The image guidance device 031 may include: one or more sets of image acquisition components, each set of image acquisition components may include a detector 0311 and an X-ray tube 0312 arranged opposite to each other. Figure 1 (Only a set of image acquisition components is shown schematically.) Among them, the X-ray tube 0312 can emit rays, such as a conical beam, and the detector 0311 can receive the rays emitted by the X-ray tube 0312. The image guiding device 031 can generate a two-dimensional projection image of the target object based on the rays received by the detector 0311.

[0041] Optionally, the X-ray tube 0312 can be a tube capable of emitting kilovolt (KV) level X-rays, and the detector 0311 can be a flat panel detector. Accordingly, the two-dimensional projection image acquired by the image guiding device 031 can be a KV level X-ray projection image.

[0042] Continue to refer to Figure 1 The radiotherapy device 04 can be equipped with multiple treatment sources 041, each of which can emit rays. The rays emitted by these multiple treatment sources 041 can rotate around a rotation axis and be focused at the beam focal point to treat the target object. Optionally, the treatment source 041 can be a gamma-ray source, and the rays emitted by the treatment source 041 are gamma rays; or, the treatment source 041 can be an X-ray source, and the rays emitted by the treatment source 041 are X-rays. Furthermore, the image guiding device 031 described in the above embodiment can also be provided on the radiotherapy device 04.

[0043] During radiotherapy, the principle behind the real-time image-guided system 03's use of IGRT technology to track the target object is as follows:

[0044] An image-guided device is used to acquire a reference image and a two-dimensional projection image of the target object, and then the two images are registered. Registering two images means using one image as the reference image and the other as the image to be registered. The purpose of registration is to ensure that all points in the image to be registered are consistent with those in the reference image. Since the two-dimensional projection image of the target object is generally an image acquired in real time during radiotherapy, i.e., an image acquired on-site, the two-dimensional projection image of the target object can be used as the image to be registered.

[0045] In related technologies, the reference image is generally a digitally reconstructed radio (DRR) image generated based on the CT image of the target object. However, since CT images are usually taken before radiotherapy to formulate a treatment plan, i.e., before positioning, and because the patient is usually positioned before generating the DRR image, errors are introduced, resulting in low accuracy of the DRR image generated from the CT image, and consequently, low registration accuracy. In this embodiment, the reference image used to track the position of the target object is an image determined based on the image acquired by the image guidance device 031 after positioning is completed, such as a DRR image generated by reconstructing the CBCT image of the target object. This avoids the positioning errors introduced when generating DRR images from CT images in related technologies, resulting in higher accuracy of the generated DRR image, and correspondingly, higher registration accuracy.

[0046] Furthermore, in order to obtain the three-dimensional offset of the target object, related technologies require first acquiring two two-dimensional offsets registered at different shooting angles. If only one set of image acquisition components is included, it is necessary to control this set of image acquisition components to first acquire a two-dimensional projection image of the target object at one angle, and then acquire another two-dimensional projection image of the target object at another angle. There is always a time interval between these two acquisitions, and the patient may move during this time interval. Therefore, not only is the real-time performance poor, but the two acquired two-dimensional projection images are also difficult to acquire in the same state, resulting in poor consistency and consequently poor registration accuracy. Including two sets of image acquisition components would occupy a large space, thus reducing the treatment space. However, in the embodiments of this disclosure, only a two-dimensional projection image of the target object at one angle can be acquired to obtain the three-dimensional offset of the target object. In other words, the image guidance device 031 only needs to include one set of image acquisition components. This ensures real-time tracking of the target object's position without affecting the treatment space and guarantees high registration accuracy.

[0047] Figure 2 This is a flowchart of a real-time image guidance method provided in an embodiment of this disclosure. This method can be applied to... Figure 1 In the real-time image guidance system 03 shown. For example... Figure 2 As shown, the method may include:

[0048] Step 201: Obtain the target reference image of the target object.

[0049] The target reference image is the image determined based on the image acquired by the image guidance device after the target object has been positioned.

[0050] Optionally, the target reference image can be a three-dimensional image or a two-dimensional image.

[0051] Step 202: Use an image guidance device to acquire a real-time projected image of the target object.

[0052] Optionally, the real-time projection image can be a two-dimensional projection image of the target object obtained by an image guidance device at the target shooting angle during radiotherapy.

[0053] Step 203: Guide the target object in real time based on the target reference image and the real-time projected image.

[0054] Optionally, the image guidance system can perform image registration based on the target reference image and the real-time projected image to obtain the target offset of the target object, and perform real-time image guidance on the target object based on the determined target offset, such as reliably adjusting the position of the target object.

[0055] In summary, the embodiments of this disclosure provide a real-time image guidance method. This image guidance system can reliably guide a target object in real-time based on an acquired target reference image and a real-time projected image. Since the target reference image is acquired after the object is positioned, the accuracy of real-time image guidance based on this target reference image and the real-time projected image is high.

[0056] Optionally, the target reference image acquired by the real-time image guidance system can be a three-dimensional image, or it can be a two-dimensional image. The following embodiments illustrate these two methods respectively:

[0057] As an alternative implementation: the target reference image is a three-dimensional image. Figure 3 This is a flowchart of another real-time image guidance method provided in this disclosure embodiment, such as... Figure 3 As shown, the method may include:

[0058] Step 301: Once the target object is positioned, the three-dimensional image of the target object reconstructed from the two-dimensional projection images of the target object under different shooting angles is determined as the target reference image.

[0059] The two-dimensional projection images of the target object at different shooting angles can be acquired by an image guidance device. Optionally, the three-dimensional image of the target can be a CBCT image.

[0060] The target reference image can be determined in the following two ways:

[0061] In the first method, after the target object is positioned, the real-time image guidance system can first use an image guidance device to re-capture two-dimensional images from different angles and reconstruct a target reference image. That is, a reference... Figure 4 The flowchart shown includes step 301, which may include:

[0062] Step 3011: After the target object is positioned, use an image guidance device to acquire two-dimensional projection images of the target object at different shooting angles.

[0063] For example, if the image guidance device includes a set of image acquisition components, it can control the X-ray tubes in these components to emit rays at different shooting angles. Correspondingly, the detector can receive these rays at different shooting angles, and the image guidance device can generate multiple two-dimensional projection images of the target at different shooting angles based on the rays received by the detector at each angle. Alternatively, the image guidance device can also include multiple sets of image acquisition components. These components can simultaneously acquire multiple two-dimensional projection images of the target at different shooting angles, thus improving the efficiency of acquiring such images.

[0064] Optionally, to ensure the quality of the target reference image obtained in subsequent reconstruction, the real-time image guidance system can use an image guidance device to perform a full-circle scan within the radiotherapy equipment 04 (i.e., scan one circle) to obtain more two-dimensional projection images of the target from different shooting angles.

[0065] In addition, the real-time image guidance system can acquire two-dimensional projection images of the target object from different shooting angles by using an image guidance device when it receives an imaging command sent by the host.

[0066] Step 3012: Reconstruct the two-dimensional projection images of the target object under different shooting angles.

[0067] After acquiring two-dimensional projection images of the target object from different shooting angles, the real-time image guidance system can reconstruct and generate a three-dimensional image of the target object based on these multiple two-dimensional projection images.

[0068] Step 3013: Determine the reconstructed 3D image of the target as the target reference image.

[0069] Finally, the real-time image guidance system can determine the reconstructed 3D image of the target as the target reference image.

[0070] It should be noted that in the first method, the target object can be positioned by registering two-dimensional images together, or by registering three-dimensional images together.

[0071] Taking the placement of the target object using two-dimensional images and two-dimensional image registration as an example, the method may further include:

[0072] Step A1: Use an image guiding device to acquire alternative two-dimensional projection images of the target object from at least two shooting angles.

[0073] Optionally, during patient positioning, the real-time image guidance system can use an image guidance device to capture images of the target object from at least two shooting angles to obtain alternative two-dimensional projection images of the target object from at least two shooting angles. The acquisition method can be referred to the description in step 3011 above, and will not be repeated here.

[0074] Step A2: Perform image registration between the candidate two-dimensional projection images at at least two shooting angles and the planned digitally reconstructed radiographic DRR images reconstructed based on the planned images at at least two shooting angles.

[0075] Optionally, the planning image can be an image obtained by scanning the target object using a planning image acquisition device when formulating a treatment plan before radiotherapy. The real-time image guidance system can acquire the planning image sent by the planning image acquisition device. For example, the planning image acquisition device can send the acquired planning image to the real-time image guidance system after receiving an image acquisition command from the real-time image guidance system. Alternatively, the planning image acquisition device can send the acquired planning image to the real-time image guidance system upon receiving an image transmission command from the host computer.

[0076] Optionally, the planned image can be a CT image or an magnetic resonance (MR) image. That is, the planned image acquisition device described above can be a CT device or an MR device. However, since both CT and MR images are three-dimensional images, in order to register with the two-dimensional projection image obtained in step A1, the real-time image guidance system can first reconstruct and generate DRR images at at least two shooting angles based on the acquired planned image. And the at least two shooting angles are the same in magnitude as the at least two shooting angles in step A1.

[0077] Then, the real-time image guidance system can perform image registration between the candidate 2D projection images obtained in step A1 at each shooting angle and the planned DRR image reconstructed from the planned image at the corresponding shooting angle to determine whether the positioning is complete. Optionally, taking a shooting angle as an example, the real-time image guidance system can use the planned DRR image at that shooting angle as the reference image and the candidate 2D projection image at that shooting angle as the image to be registered, compare the coordinates of each point in the two images, and determine whether the positioning is complete based on the comparison result.

[0078] Step A3: Once the registration result meets the registration conditions, the target object is positioned.

[0079] Optionally, the registration condition can be: the positional deviation of the target object in the two registered images is less than or equal to the deviation threshold. For example, if the deviation threshold can be 0, and the registration condition is that the positional deviation of the target object in the two registered images is equal to the deviation threshold, then the registration result satisfying the registration condition can mean that the position of the target object in the two registered images is completely consistent.

[0080] It should be noted that in the application scenarios described in the embodiments of this disclosure, the deviation is generally a three-dimensional deviation, and the above is only an illustrative description.

[0081] If, after executing step A2, the registration result of the candidate two-dimensional projection image at each shooting angle and the planned digitally reconstructed radiographic DRR image reconstructed based on the planned image at the corresponding shooting angle meets the above registration conditions, then the real-time image guidance system can determine that the positioning is complete. Then, the candidate two-dimensional projection images at at least two shooting angles obtained in step A1 can be further determined as target two-dimensional projection images at different shooting angles for execution in step 3011. That is, after executing step A3, step 3011 can be executed immediately thereafter.

[0082] Of course, if the registration result does not meet the registration conditions, steps A1 to A3 can be continued until the registration result meets the registration conditions.

[0083] In the second approach, the reconstructed 3D image used for registration when the target object is positioned is directly determined as the target reference image. Therefore, before obtaining the target reference image of the target object, the method may further include:

[0084] Step B1: Use an image guidance device to acquire alternative 3D images of the target object.

[0085] The alternative 3D image can be an image reconstructed based on a reference 2D projection image of the target object at different shooting angles.

[0086] That is, during the positioning process, the real-time image guidance system can first use an image guidance device to acquire reference two-dimensional projection images of the target object at different shooting angles, and then reconstruct the reference two-dimensional projection images of the target object at different shooting angles to obtain alternative three-dimensional images of the target object.

[0087] Optionally, the method for acquiring reference two-dimensional projection images of the target object at different shooting angles using an image guidance device can refer to step 3011, and will not be repeated here. Furthermore, the alternative three-dimensional image can be a CBCT image.

[0088] Step B2: Perform image registration between the candidate 3D images and the planned images.

[0089] Since step B1 yields both a candidate 3D image and a planned 3D image, the real-time image guidance system can directly register the candidate and planned 3D images. The registration method can be found in the description of step A2, and will not be repeated here.

[0090] Step B3: Once the registration result meets the registration conditions, the target object is positioned.

[0091] Similar to step A3, if the registration result meets the registration conditions, the real-time image guidance system can determine that the positioning is complete and can continue to execute step B4 below. Of course, if the registration result does not meet the registration conditions, steps B1 to B3 can be continued until the registration result meets the registration conditions.

[0092] Step B4: Select the candidate 3D image as the target reference image (Step B4 is another specific example of step 301).

[0093] Finally, the real-time image guidance system can determine the candidate 3D image used for registration as the target 3D image. That is, through steps B1 to B4, the target reference image after the positioning is completed can be directly determined. It should be noted that in the second method, steps B1 to B3 are to complete the positioning of the target object by using 3D images and 3D image registration.

[0094] Step 302: Use an image guidance device to acquire a real-time projection image of the target object.

[0095] As described in the above embodiments, once the target reference image of the target object is obtained, the positioning is considered complete. At this point, the patient can be sent into the treatment space of the radiotherapy equipment to receive radiotherapy.

[0096] Although the patient is placed in the treatment space only after positioning is complete, the target's position may shift during radiotherapy due to unavoidable factors such as patient movement, breathing, or coughing. Therefore, to ensure treatment accuracy and avoid accidental irradiation of other normal tissues, image-guided technology is needed to track the target's position in real time during radiotherapy. This allows for real-time adjustment of the patient's position, ensuring that the treatment beam focus is aligned with the target's treatment point. To track the target's position, the real-time image-guided system first needs to acquire the target's current location.

[0097] For example, a real-time image guidance system can use an image guidance device to acquire a real-time projected image of the target object at the target shooting angle, i.e., a two-dimensional projected image of the target object. The acquisition method can be referred to the embodiments described above. Furthermore, the real-time image guidance system can begin acquiring a real-time projected image of the target object using the image guidance device after receiving an imaging command sent by the host.

[0098] Step 303: Obtain the target shooting angle of the real-time projected image.

[0099] To facilitate subsequent image registration, the real-time image guidance system can also obtain the shooting angle of the image guidance device when acquiring real-time projected images, i.e., the target shooting angle.

[0100] Step 304: Obtain the target's three-dimensional image and the target's digitally reconstructed radiographic (DRR) image at the target's shooting angle.

[0101] Based on the steps described above, it can be seen that the image acquired in real-time during treatment, reflecting the current position of the target object, is a two-dimensional image, while the image acquired after positioning is a three-dimensional image. Therefore, before registration, the real-time image guidance system needs to reconstruct and generate a DRR image of the target object at the target shooting angle based on this three-dimensional image. The position of the target object in this DRR image is the reference position that the target object should be in when the target point is aligned with the focal point of the treatment beam.

[0102] It should be noted that for each 2D projection image at each shooting angle, there can be a corresponding DRR image at the same angle. That is, regardless of the target shooting angle, the real-time image guidance system can generate a DRR image at that target shooting angle based on the target 3D image of the target object.

[0103] Since the target 3D image is acquired after the setup is completed, compared with related technologies that generate DRR images based on the planned image before setup, the DRR image generated based on the target 3D image can more accurately represent the reference position of the target object and improve the registration results.

[0104] Optionally, before positioning the patient after developing a treatment plan, a marker can be placed on or inside the patient's body, such as by attaching it to the patient's skin or implanting it in the patient's body.

[0105] Optionally, the marker can be a metal marker (or simply, a gold marker). Because markers produce better images, they can improve registration accuracy during registration. Furthermore, at least three non-collinear markers can be used. This allows the real-time image guidance system to reference the positions of multiple markers at different shooting angles to register the images, further improving registration accuracy.

[0106] For example, suppose the target is a tumor located in the head, and three non-collinear markers are set. Then refer to... Figure 5 A marker B0 can be placed at each of the patient's temples and the tip of their nose. If the target is located on the body, a marker B0 can be placed along the patient's spine.

[0107] Since marker B0 is set before the setup, both the target reference image obtained after setup and the real-time projection image obtained during setup can include marker B0. To improve registration accuracy, step 304 may further include:

[0108] Image processing is performed on the target 3D image to obtain a target 3D image containing only the landmarks. Then, a target DRR image containing only the landmarks is obtained at the target shooting angle. That is, the DRR image used for subsequent registration can be a DRR image containing only the landmarks.

[0109] It should be noted that the real-time image guidance system can store at least two target reference images. One target reference image may only include the markers, and this target reference image containing only markers can be used for subsequent real-time image guidance. The other target reference image may include the markers and other information (e.g., bone tissue), and this target reference image can be used for display, such as displaying it to the treating physician via a host computer.

[0110] Step 305: Based on the target DRR image and the real-time projection image, perform real-time image guidance on the target object.

[0111] Optional, as an alternative implementation method, see reference. Figure 6 The flowchart shown includes step 305, which may include:

[0112] Step 3051A: Perform image registration on the target DRR image and the real-time projection image.

[0113] After acquiring the target DRR image and the real-time projection image, the real-time image guidance system can use the target DRR image as the reference image and the real-time projection image as the image to be registered, and compare the coordinates of each point in the two images to determine the offset of the target object.

[0114] An alternative embodiment: If the target object includes a marker, then step 3051A may include:

[0115] The markers in the target DRR image and the real-time projection image are segmented separately, and the markers in the segmented target DRR image and the real-time projection image are image registered.

[0116] Optionally, the segmentation method can be as follows: The real-time image guidance system can first perform image blurring on the image to be segmented (e.g., the target DRR image and the real-time projection image) to blur the markers in the image to be segmented, making the markers blend into the background of the image to be segmented, thus obtaining a new image. Then, the real-time image guidance system can use the image to be segmented minus the new image obtained after image blurring to complete the segmentation of the markers. Image subtraction can refer to: subtracting the pixel value of a pixel in the image to be segmented from the pixel value of the new image obtained after blurring.

[0117] Alternatively, the real-time image guidance system can first acquire the CT values ​​of each target object and marker in the image to be segmented. The CT value, measured in Hu (Human Units), measures the absorption rate of radiation by human tissue. The system can also preset a threshold for CT filtering (referred to as a reference threshold). Then, based on the CT values ​​of each target object and marker, and the reference threshold, the system can perform image normalization processing on the image to be segmented, completing the marker segmentation. For example, the system can determine the relationship between the CT values ​​of each target object and marker and the reference threshold. For target objects with CT values ​​greater than the reference threshold, the system does not change their CT values. For target objects with CT values ​​less than the reference threshold, the system can set the CT value as the first threshold and the CT value of the marker as the second threshold.

[0118] Alternatively, the target shooting angle of the image to be segmented can be obtained, the reconstructed 3D image can be obtained, and the target DRR image of the reconstructed 3D image at the target shooting angle can be obtained. Optionally, the reconstructed 3D image can be filtered to obtain a 3D image containing only the landmarks, and the target DRR image of the 3D image containing only the landmarks at the target shooting angle can be obtained. Then, one or more reference regions of interest (ROIs) are constructed in the target DRR image. Each reference ROI contains one or more landmarks, and the one or more reference ROIs are mapped onto the image to be segmented to obtain one or more reference target ROIs, thus completing the segmentation.

[0119] Optionally, image registration of the markers in the segmented target DRR image and the real-time projection image can be performed by comparing the positions of the target point of the marker in the target DRR image and the real-time projection image. This target point can be the center point of the marker.

[0120] Another optional embodiment: If the target object includes bony landmark tissues and markers, then step 3051A may include: segmenting the markers in the target DRR image and the real-time projection image respectively, performing a first target registration on the segmented markers in the target DRR image and the markers in the real-time projection image; and performing a second target registration on the bony landmark tissues in the target DRR image and the bony landmark tissues in the real-time projection image. The comparison method can be referred to the above embodiments, and will not be repeated here.

[0121] Step 3052A: Determine the target offset of the target object based on the registration result.

[0122] Optionally, if image registration is performed solely based on markers, the real-time image guidance system can directly determine the offset obtained from registration as the target offset of the target object.

[0123] Optionally, if image registration is performed based on both markers and bony landmarks, the real-time image guidance system can determine the first reference offset of the target object based on the registration result of the first target registration, determine the second reference offset of the target object based on the registration result of the second target registration, and calculate the target offset of the target object based on the first offset, the weight value of the first offset, the second offset, and the weight value of the second offset.

[0124] Optionally, the real-time image guidance system can have preset weight values ​​for a first offset and a second offset. After obtaining the first and second offsets, the image guidance system can use the first offset, its weight value, the second offset, and its weight value as parameters to calculate the actual offset of the target object through a weighted summation. Combining these two methods to determine the target object's offset offers both good reliability and accuracy.

[0125] It should be noted that if the real-time image guidance system cannot obtain the offset based on the image at a certain shooting angle, it can choose not to output the result to avoid the risk of erroneous output.

[0126] Step 3053A: Adjust the position of the target object according to the target offset.

[0127] Finally, the real-time image-guided system can flexibly adjust the position of the target object based on the target offset, ensuring treatment accuracy. For example, after registering the target offset, the real-time image-guided system can send the target offset to the host computer, allowing the host computer to flexibly adjust the patient's position in real time, ensuring treatment accuracy.

[0128] Optionally, since the target offset may have a negligible impact on the accuracy of radiotherapy when it is small, the real-time image guidance system can first detect whether the target offset is greater than an offset threshold after obtaining it, that is, detect whether the patient's movement deviation is large. If it is greater than the offset threshold, the real-time image guidance system can further output the offset to the host for position adjustment. Alternatively, the host can detect whether the target offset is greater than the offset threshold, and then position the target object if it is greater than the offset threshold. This disclosure does not limit this aspect.

[0129] Optionally, as another possible implementation, see [reference]. Figure 7 The flowchart shown includes step 305, which may include:

[0130] Step 3051B: Determine the image scaling factor based on the target DRR image and the real-time projected image.

[0131] Optionally, the real-time image guidance system can determine the image scaling factor of the real-time projected image by comparing the target DRR image and the real-time projected image. In this method, only a real-time projected image from a single shooting angle needs to be acquired.

[0132] For example, taking a shooting angle as an example, if the target object moves downwards at that shooting angle, the current real-time projected image at that angle is equivalent to a scaled-down image before it moved downwards. If the target object moves upwards at that shooting angle, the current real-time projected image at that angle is equivalent to a scaled-up image before it moved upwards. In this way, the image scaling factor can be determined.

[0133] Optionally, if the target object also includes markers, then step 3051B may include: segmenting the markers in the target DRR image and the real-time projection image respectively, and determining the image scaling factor based on the markers in the segmented target DRR image and the markers in the real-time projection image.

[0134] Step 3052B: Determine the target offset of the target object based on the image scaling factor.

[0135] Once the image scaling factor is determined, the real-time image guidance system can determine the offset of the target object in the contraction direction based on that scaling factor. Then, by combining this offset with two other offsets determined in the two-dimensional coordinate system, the target offset of the target object can be obtained.

[0136] Step 3053B: Adjust the position of the target object according to the target offset.

[0137] This step can be referred to as step 3053A above, and will not be repeated here.

[0138] As an alternative implementation, the target reference image is a two-dimensional image. Figure 8 This is a flowchart of another real-time image guidance method provided in this disclosure embodiment, such as... Figure 8 As shown, the method may include:

[0139] Step 401: Once the target object is positioned, determine the two-dimensional projection images of the target object at different shooting angles as the target reference images.

[0140] Specifically, the two-dimensional projection images of the target object at different shooting angles can be acquired by the image guidance device. That is, the real-time image guidance system can acquire the two-dimensional projection images of the target object at different shooting angles using the image guidance device.

[0141] Step 402: Use an image guidance device to acquire a real-time projection image of the target object.

[0142] This step can be referred to in the description of step 302 above, and will not be repeated here.

[0143] Step 403: Obtain the target shooting angle of the real-time projected image.

[0144] This step can be referred to in the description of step 303 above, and will not be repeated here.

[0145] Step 404: Obtain the two-dimensional projection image of the target object at the target shooting angle.

[0146] After determining the target shooting angle of the acquired real-time projection image, the real-time image guidance system can further determine the target two-dimensional projection image at that shooting angle from the target two-dimensional projection images acquired at different shooting angles.

[0147] Step 405: Determine the image scaling factor based on the two-dimensional projection image of the target at the target shooting angle and the real-time projection image.

[0148] This step can be referred to in the description of step 3051B above, and will not be repeated here.

[0149] Similarly, if the target object includes a marker, then step 405 may include:

[0150] The markers in the target 2D projection image and the real-time projection image at the target shooting angle are segmented respectively, and the image scaling factor is determined based on the segmented markers in the target 2D projection image at the target shooting angle and the real-time projection image.

[0151] Step 406: Determine the target offset of the target object based on the image scaling factor.

[0152] This step can be referred to in the description of step 3052B above, and will not be repeated here.

[0153] Step 407: Adjust the position of the target object according to the target offset.

[0154] This step can be referred to in the description of step 3053B above, and will not be repeated here.

[0155] Optionally, after the target object has been positioned, the real-time image guidance system can also first obtain the positioning and registration type of the target object during the positioning stage, and flexibly determine the target reference image based on the positioning and registration type. That is, such as... Figure 9 Another real-time image guidance method is shown. For example... Figure 9 As shown, the method may include:

[0156] Step 501: Once the target object has been positioned, obtain the positioning registration type of the target object.

[0157] Optionally, in this embodiment of the present disclosure, after the target object has been positioned, the real-time image guidance system can first obtain the positioning and registration type of the target object. This positioning and registration type can be the type of image registration during the positioning stage. Based on the above embodiments, this positioning and registration type can be 2D-2D registration or 3D-3D registration.

[0158] Step 502: Determine the target reference image based on the image acquired by the image guidance device according to the positioning and registration type.

[0159] If the positioning and registration type package is 2D-2D registration, then in accordance with the above embodiments, the real-time image guidance system can first use an image guidance device to acquire two-dimensional projection images of the target object at different shooting angles, then reconstruct the two-dimensional projection images of the target object at different shooting angles, and determine the reconstructed three-dimensional image of the target as the target reference image.

[0160] If the positioning and registration type package is 3D-3D registration, then in accordance with the above embodiments, the real-time image guidance system can directly determine the target three-dimensional image that is registered with the planned image when the positioning is completed as the target reference image.

[0161] Step 503: Use an image guidance device to acquire a real-time projected image of the target object.

[0162] This step can be referred to in the description of step 302 above, and will not be repeated here.

[0163] Step 504: Guide the target object in real time based on the target reference image and the real-time projected image.

[0164] Optionally, if the target object includes a marker, step 504 may include:

[0165] The markers in the target reference image and the real-time projected image are segmented separately. A first target registration is performed between the segmented markers in the target reference image and the real-time projected image, and the target offset of the target object is determined based on the registration result of the first target registration. Furthermore, the position of the target object can be adjusted according to the target offset to achieve real-time image guidance of the target object.

[0166] Optionally, if the target object includes bony landmark tissues and markers, step 504 may include:

[0167] Landmarks in both the target reference image and the real-time projection image are segmented. A first target registration is performed between the segmented landmarks in the target reference image and the real-time projection image to determine a first reference offset for the target object. Then, a second target registration is performed between bony landmarks in the target reference image and the real-time projection image, and a second reference offset for the target object is determined based on the second target registration result. Finally, the target offset for the target object is calculated based on the first offset, its weight, the second offset, and its weight. Furthermore, the position of the target object can be adjusted according to the target offset to achieve real-time image guidance of the target object.

[0168] The optional implementation methods of the above steps can be referred to the corresponding embodiments described above, and will not be repeated here.

[0169] It should be noted that the order of steps in the real-time image guidance method provided in this disclosure can be appropriately adjusted. Any variations that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this disclosure should be included within the protection scope of this disclosure, and therefore will not be elaborated further.

[0170] In summary, the embodiments of this disclosure provide a real-time image guidance method. This image guidance system can reliably guide a target object in real-time based on an acquired target reference image and a real-time projected image. Since the target reference image is acquired after the object is positioned, the accuracy of real-time image guidance based on this target reference image and the real-time projected image is high.

[0171] Figure 10 This is a block diagram of a real-time image guidance device provided in an embodiment of this disclosure. This device can be applied to... Figure 1 In the real-time image guidance system 03 shown. For example... Figure 10As shown, the device may include:

[0172] The first acquisition module 601 is used to acquire the target reference image of the target object.

[0173] The target reference image can be an image determined based on the image acquired by the image guidance device after the target object has been positioned.

[0174] The second acquisition module 602 is used to acquire a real-time projected image of the target object using an image guidance device.

[0175] The image guidance module 603 is used to guide the target object in real time based on the target reference image and the real-time projected image.

[0176] Optionally, as an alternative implementation: the target reference image is a three-dimensional image. Then, the first acquisition module 601 can be used for:

[0177] Once the target object is positioned, the reconstructed 3D image of the target object from the 2D projection images of the target object at different shooting angles is determined as the target reference image. The 2D projection images of the target object at different shooting angles are acquired by an image guiding device.

[0178] That is, the first acquisition module 601 can be used for:

[0179] Once the target object is positioned, an image guidance device is used to acquire two-dimensional projection images of the target object from different shooting angles.

[0180] Reconstruct two-dimensional projection images of the target object from different shooting angles.

[0181] The reconstructed 3D image of the target is used as the target reference image.

[0182] Optional, such as Figure 11 As shown, the device may further include:

[0183] The fourth acquisition module 604 is used to acquire alternative three-dimensional images of the target object using an image guidance device before acquiring the target reference image of the target object. The alternative three-dimensional images are images reconstructed based on the reference two-dimensional projection images of the target object under different shooting angles.

[0184] The second image registration module 605 is used to perform image registration between the candidate 3D image and the planned image.

[0185] The second determining module 606 is used to determine that the target object is positioned when the registration result meets the registration conditions.

[0186] Accordingly, the first acquisition module 601 can be used to determine the candidate 3D image as the target reference image.

[0187] Optionally, as an alternative implementation: the image guidance module 603 can be used for:

[0188] Obtain the target shooting angle of the real-time projected image.

[0189] Obtain the target's 3D image and the target's digital reconstructed radiographic (DRR) image at the target's shooting angle.

[0190] Real-time image guidance is provided for the target object based on the target DRR image and the real-time projected image.

[0191] For example, the image guidance module 603 can be used for:

[0192] Image registration is performed on the target DRR image and the real-time projection image. Based on the registration results, the target offset of the target object is determined, and the position of the target object is adjusted according to the target offset.

[0193] Optionally, the target object may include a marker, and accordingly, the image guidance module 603 may be used for:

[0194] The markers in the target DRR image and the real-time projection image are segmented separately, and the markers in the segmented target DRR image and the real-time projection image are image registered.

[0195] Optionally, the image guidance module 603 can be used to: perform image processing on the target 3D image to obtain a target 3D image that includes only the markers, and acquire a target DRR image of the target 3D image that includes only the markers at the target shooting angle.

[0196] Alternatively, as another optional implementation, the image guidance module 603 can be used for:

[0197] The image scaling factor is determined based on the target DRR image and the real-time projected image.

[0198] The target offset of the target object is determined based on the image scaling factor.

[0199] Adjust the position of the target object based on the target offset.

[0200] Optionally, if the target object includes landmarks, the image guidance module 603 can be used to:

[0201] The markers in the target DRR image and the real-time projection image are segmented separately, and the image scaling factor is determined based on the markers in the segmented target DRR image and the real-time projection image.

[0202] Optionally, as another possible implementation: if the target reference image is a two-dimensional image, then the first acquisition module 601 can be used to:

[0203] Once the target object is positioned, the two-dimensional projection images of the target object at different shooting angles are determined as target reference images. These two-dimensional projection images of the target object at different shooting angles are acquired by an image guiding device.

[0204] Optionally, the image guidance module 603 can be used to: acquire the target shooting angle of the real-time projected image; acquire the target two-dimensional projected image of the target object at the target shooting angle; determine the image scaling factor based on the target two-dimensional projected image at the target shooting angle and the real-time projected image; determine the target offset of the target object based on the image scaling factor; and adjust the position of the target object based on the target offset.

[0205] Optionally, if the target object includes a landmark, the image guidance module 603 can be used to:

[0206] The markers in the target 2D projection image and the real-time projection image at the target shooting angle are segmented respectively, and the image scaling factor is determined based on the segmented markers in the target 2D projection image at the target shooting angle and the real-time projection image.

[0207] Optionally, the marker may include a metal marker, which may be affixed to the patient's skin or implanted in the patient during the placement phase.

[0208] Optionally, the target object may include at least three non-collinear markers.

[0209] Alternatively, as yet another optional implementation: the first acquisition module 601 can be used for:

[0210] Once the target object is positioned, the positioning and registration type of the target object is obtained. The positioning and registration type is the type of image registration during the positioning stage. Based on the positioning and registration type, the target reference image is determined based on the image acquired by the image guidance device.

[0211] Optionally, the registration type can include 2D-2D registration and 3D-3D registration.

[0212] If the positioning and registration type includes 2D-2D registration, the first acquisition module 601 can be used to: acquire two-dimensional projection images of the target object at different shooting angles using an image guidance device, reconstruct the two-dimensional projection images of the target object at different shooting angles, and determine the reconstructed three-dimensional image of the target as the target reference image.

[0213] If the positioning and registration type includes 3D-3D registration, the first acquisition module 601 can be used to: determine the target three-dimensional image that is registered with the planned image when the positioning is completed as the target reference image.

[0214] In summary, the embodiments of this disclosure provide a real-time image guidance device. This device can reliably guide a target object in real-time based on an acquired target reference image and a real-time projected image. Since the target reference image is acquired after the object is positioned, the accuracy of real-time image guidance based on this target reference image and the real-time projected image is high.

[0215] Regarding the real-time image guidance device in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated here.

[0216] Optional, combined Figure 1 and Figure 12 The real-time image guidance system 03 in a radiotherapy system may include: an image guidance device 031, a processor 032, and a memory 033. The image guidance device is used to acquire images. The memory may store instructions. These instructions are loaded and executed by the processor to achieve, for example... Figure 3 , Figure 8 and Figure 9 Any of the real-time image guidance methods shown.

[0217] Optionally, embodiments of this disclosure also provide a storage medium that may store instructions, which, when the storage medium is running on a processing component, cause the processing component to execute instructions such as 3. Figure 8 and Figure 9 Any of the real-time image guidance methods shown.

[0218] The above description is only an optional embodiment of this disclosure and is not intended to limit this disclosure. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this disclosure should be included within the protection scope of the embodiments of this disclosure.

Claims

1. A real-time image guidance method, characterized in that, The method includes: During the positioning process, an image guidance device is used to acquire reference two-dimensional projection images of the target object at different shooting angles; the reference two-dimensional projection images of the target object at different shooting angles are reconstructed to obtain candidate three-dimensional images of the target object; image registration is performed between the candidate three-dimensional images and the planned images, where the planned images are three-dimensional images; when the registration result meets the registration conditions, the candidate three-dimensional images are determined as the target reference images, and the positioning of the target object is completed; wherein, the target object includes bony landmarks and metallic markers, and the metallic markers are applied to the patient's body surface or implanted in the patient's body after the treatment plan is formulated and before positioning; During radiotherapy, the image guidance device acquires a real-time projection image of the target object at the target imaging angle; the real-time projection image is a two-dimensional projection image of the target object. The target imaging angle at which the image guidance device acquires the real-time projection image is located is also acquired. A digitally reconstructed radiographic refraction (DRR) image of the target object at the target imaging angle is acquired. Metallic markers in the target DRR image and the real-time projection image are segmented, and a first target registration is performed on the segmented metallic markers in the target DRR image and the real-time projection image. Finally, the target DRR image is reconstructed. A second target registration is performed between bony landmarks in the first target registration image and bony landmarks in the real-time projection image; a first offset of the target object is determined based on the registration result of the first target registration, and a second offset of the target object is determined based on the registration result of the second target registration; a target offset of the target object is calculated based on the first offset, the weight value of the first offset, the second offset, and the weight value of the second offset; if the target offset is greater than an offset threshold, the position of the target object is adjusted according to the target offset; if the offset cannot be obtained based on the image at a certain shooting angle, no result is output; The segmentation of the metallic markers in the target DRR image and the real-time projection image includes: blurring the image to be segmented to remove the metallic markers so that they blend into the background of the image to be segmented, resulting in a new image; and subtracting the new image obtained after blurring the image from the image to be segmented to complete the segmentation of the metallic markers. Image subtraction refers to subtracting the pixel values ​​of the pixels in the new image obtained after blurring the image from the pixel values ​​of the pixels in the image to be segmented. The image to be segmented includes the DRR image and the real-time projection image.

2. The method according to claim 1, characterized in that, The target objects include at least three non-collinear metallic markers.

3. A real-time image guidance system, characterized in that, The real-time image guidance system includes: an image guidance device, a processor, and a memory; The image guidance device is used to acquire images, and the memory stores instructions, which are loaded and executed by the processor to implement the real-time image guidance method as described in claim 1 or 2.

4. The system according to claim 3, characterized in that, The image guidance device is a target cone-beam electro-optical CT (CBCT) device.

5. A storage medium, characterized in that, The storage medium stores instructions that, when the storage medium is running on the processing component, cause the processing component to execute the real-time image guidance method as described in claim 1 or 2.

6. A radiotherapy system, characterized in that, The radiotherapy system includes: a patient support device, a main unit, and a real-time image guidance system; the real-time image guidance system is the system as described in claim 3 or 4. The host is connected to both the real-time image guidance system and the patient support device. The real-time image guidance system sends the target offset of the determined target object to the host, and the host adjusts the position of the patient support device based on the target offset.

Images