A Geometric Correction Method and System Based on Image Tracking

Abstract

A geometric correction method and system based on image tracking. The method includes: obtaining multiple projection images of a geometric phantom, where each projection sphere in the multiple projection images has a corresponding relationship with each original sphere in the geometric phantom, and the multiple projection images include a first projection image at a first angle; obtaining the corresponding relationship between each first projection sphere and each original sphere in the first projection image; tracking the projection spheres based on the similarity between projection images at adjacent angles to obtain a tracking result; obtaining the corresponding relationship between each original sphere and each projection sphere based on the tracking result; and performing geometric correction based on the corresponding relationship.

Description

Technical Field

[0001] This specification relates to the field of medical technology, and particularly to a geometric correction method and system based on image tracking. Background Art

[0002] Geometric correction is one of the necessary corrections for image reconstruction. A special geometric correction phantom is required. By taking pictures of the geometric phantom at various rotation angles, the actual spatial coordinates of the pixels captured by the detector at each angle are calculated. In the calculation of geometric correction, an important step is to determine the correspondence between each small ball in the geometric phantom and the projection pixel coordinates in the captured projection image. This correspondence is detected and calculated separately for each projection image, and the requirements for the phantom positioning are very strict. If the initial angle deviation of the positioning is too large, it will cause calculation errors.

[0003] Based on this, there is an urgent need for an efficient and accurate geometric correction method. Summary of the Invention

[0004] One embodiment of this specification provides a geometric correction method based on image tracking. The method includes: obtaining multiple projection images of a geometric phantom, where each projection small ball in the multiple projection images has a correspondence with each original small ball in the geometric phantom, and the multiple projection images include a first projection image at a first angle; obtaining the correspondence between each first projection small ball and each original small ball in the first projection image; tracking the projection small balls based on the similarity between the projection images at adjacent angles to obtain a tracking result; obtaining the correspondence between each original small ball and each projection small ball based on the tracking result; and performing geometric correction based on the correspondence.

[0005] One embodiment of this specification provides a geometric correction system based on image tracking, including: an obtaining module, configured to obtain multiple projection images of a geometric phantom, where each projection small ball in the multiple projection images has a correspondence with each original small ball in the geometric phantom, and the multiple projection images include a first projection image at a first angle; obtaining the correspondence between each first projection small ball and each original small ball in the first projection image; a tracking module, configured to track the projection small balls based on the similarity between the projection images at adjacent angles to obtain a tracking result; a correspondence determination module, configured to obtain the correspondence between each original small ball and each projection small ball based on the tracking result; and a correction module, configured to perform geometric correction based on the correspondence.

[0006] One embodiment of this specification provides a geometric correction device based on image tracking. The device includes: a display device, configured to display at least one of a projection image of a geometric phantom and the correspondence between each small ball in the projection image; at least one storage medium, configured to store computer instructions; and at least one processor, configured to execute the computer instructions to implement the geometric correction method based on image tracking as described above.

[0007] One embodiment of this specification provides a computer-readable storage medium. The storage medium stores computer instructions. When a computer reads the computer instructions, the computer executes the geometric correction method based on image tracking as described above.

[0008] In the prior art, a method of manually positioning a phantom and separately detecting and calculating the correspondence between small balls and projections at each positioning angle is usually adopted. This method has high precision requirements for phantom positioning and low efficiency.

[0009] An embodiment of this specification proposes a method of automatically selecting specific angles and using the relationship between projections at adjacent shooting angles to track small balls to achieve the calculation purpose. This method does not require very precise positioning angles, greatly reducing the difficulty of geometric correction positioning and improving the efficiency of the correction work. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] This specification will be further described by way of exemplary embodiments, which will be described in detail through the accompanying drawings. These embodiments are not restrictive. In these embodiments, the same numbers represent the same structures, where:

[0011] Figure 1 is a schematic diagram of an application scenario of an exemplary geometric correction system based on image tracking shown in some embodiments of this specification;

[0012] Figure 2 is a block diagram of an exemplary geometric correction system based on image tracking shown in some embodiments of this specification;

[0013] Figure 3 is a flowchart of an exemplary geometric correction method based on image tracking shown in some embodiments of this specification;

[0014] Figure 4 is a schematic diagram of a projection map shown in some embodiments of this specification;

[0015] Figure 5 is a schematic diagram of a geometric correction method based on image tracking shown in some embodiments of this specification;

[0016] Figure 6 is a schematic diagram of an identification model and its training shown in some embodiments of this specification. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] To more clearly illustrate the technical solutions of the embodiments of this specification, the following provides a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings in the following description are merely some examples or embodiments of this specification. For those of ordinary skill in the art, without creative efforts, this specification can also be applied to other similar scenarios based on these drawings. Unless obvious from the language context or otherwise stated, the same reference numerals in the figures represent the same structure or operation.

[0018] It should be understood that the "system", "device", "unit", and / or "module" used herein is a way to distinguish different components, elements, parts, portions, or assemblies at different levels. However, if other words can achieve the same purpose, the said words can be replaced by other expressions.

[0019] As shown in this specification and the claims, unless the context clearly indicates an exceptional situation, words such as "a", "an", "one", and / or "the" are not specifically singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of the clearly identified steps and elements, and these steps and elements do not constitute an exclusive list. A method or device may also include other steps or elements.

[0020] Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the operations before or after may not necessarily be executed precisely in sequence. On the contrary, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0021] Figure 1 is a schematic diagram of the application scenario of an exemplary image-tracking-based geometric correction system shown according to some embodiments of this specification. In some embodiments, as Figure 1 shown, the application scenario 100 of the image-tracking-based geometric correction system may at least include an imaging device 110, a processing device 120, a terminal device 130, a storage device 140, and a network 150.

[0022] The imaging device 110 can scan a target object within a detection area or a scanning area to obtain scanning data of the target object. In some embodiments, the target object may include a biological object and / or a non-biological object. For example, the target object may include a patient, an artificial object, etc. In some embodiments, the target object may include a specific part of the body, such as the head, chest, abdomen, etc. or any combination thereof. In some embodiments, the target object may include a specific organ, such as the heart, esophagus, trachea, bronchus, stomach, gallbladder, small intestine, colon, bladder, ureter, uterus, fallopian tube, etc. or any combination thereof. In some embodiments, the target object may include a region of interest (ROI), such as a tumor, a node, etc.

[0023] In some embodiments, the imaging device 110 may be or include an X-ray imaging device. For example, the X-ray imaging device may include DSA (Digital Subtraction Angiography), Digital Radiography (DR), Computed Radiography (CR), Digital Fluorography (DF), mammography machine, C-arm device, etc. In some embodiments, the imaging device 110 may include a single-modal scanner and / or a multi-modal scanner. The single-modal scanner may include, for example, a CT scanner, a magnetic resonance imaging (MRI) scanner, etc. The multi-modal scanner may include, for example, an X-ray imaging - magnetic resonance imaging (X-ray - MRI) scanner, a digital subtraction angiography - magnetic resonance imaging (DSA - MRI) scanner, etc. or any combination thereof. The above descriptions of the imaging device are for illustrative purposes only and are not intended to limit the scope of this specification.

[0024] The processing device 120 can process data and / or information obtained from the imaging device 110, the terminal device 130, the storage device 140, and / or other components of the application scenario 100 of the image-tracking-based geometric correction system. For example, the processing device 120 can obtain images (such as medical images, projection diagrams of geometric phantoms, etc.) from the terminal device 130 and the storage device 140 and perform analysis and processing on them.

[0025] In some embodiments, the processing device 120 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, the processing device 120 may be local or remote. For example, the processing device 120 may access information and / or data from the imaging device 110, the terminal device 130, and / or the storage device 140 via the network 150. Alternatively, the processing device 120 may be directly connected to the imaging device 110, the terminal device 130, and / or the storage device 140 to access information and / or data. In some embodiments, the processing device 120 may be implemented on a cloud platform. For example, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an inter-cloud, a multi-cloud, etc., or any combination thereof.

[0026] In some embodiments, the processing device 120 and the imaging device 110 may be integrated into one body. In some embodiments, the processing device 120 and the imaging device 110 may be directly or indirectly connected and cooperate to implement the methods and / or functions described in this specification.

[0027] In some embodiments, the processing device 120 may include an input device and / or an output device. Through the input device and / or the output device, interaction with the user can be achieved (e.g., setting scanning parameters, etc.). In some embodiments, the input device and / or the output device may include a display screen, a keyboard, a mouse, a microphone, etc., or any combination thereof.

[0028] The terminal device 130 may communicate with and / or be connected to the imaging device 110, the processing device 120, and / or the storage device 140. In some embodiments, interaction with the user may be achieved through the terminal device 130. In some embodiments, the terminal device 130 may include a mobile device 131, a tablet computer 132, a laptop computer 133, etc., or any combination thereof. In some embodiments, the terminal device 130 (or all or part of its functions) may be integrated into the processing device 120.

[0029] The storage device 140 may store data, instructions, and / or any other information. In some embodiments, the storage device 140 may store data obtained from the imaging device 110, the processing device 120, the terminal device 130, and / or (e.g., projection diagrams of a geometric phantom, tracking results, etc.). In some embodiments, the storage device 140 may store the data and / or instructions used by the processing device 120 to execute or use to complete the exemplary methods described in this specification.

[0030] In some embodiments, the storage device 140 may include one or more storage components, and each storage component may be an independent device or a part of other devices. In some embodiments, the storage device 140 may include a random access memory (RAM), a read-only memory (ROM), a mass storage device, a removable storage device, a volatile read-write memory, etc., or any combination thereof. In some embodiments, the storage device 140 may be implemented on a cloud platform. In some embodiments, the storage device 140 may be a part of the imaging device 110, the processing device 120, and / or the terminal device 130.

[0031] The network 150 may include any suitable network capable of facilitating information and / or data exchange. In some embodiments, at least one component of the application scenario 100 of the image-tracking-based geometric correction system (e.g., the imaging device 110, the processing device 120, the terminal device 130, the storage device 140) may exchange information and / or data with at least one other component in the application scenario 100 of the image-tracking-based geometric correction system through the network 150. For example, the processing device 120 may obtain medical images, projection maps of geometric phantoms, etc. from the imaging device 110 through the network 150.

[0032] It should be noted that the above description of the application scenario 100 of the image-tracking-based geometric correction system is provided for illustrative purposes only and is not intended to limit the scope of this specification. For those of ordinary skill in the art, various modifications or changes can be made according to the description of this specification. For example, the application scenario 100 of the image-tracking-based geometric correction system may implement similar or different functions on other devices. However, these changes and modifications will not deviate from the scope of this specification.

[0033] Figure 2 is a module diagram of an exemplary image-tracking-based geometric correction system shown according to some embodiments of this specification. As Figure 2 shown, in some embodiments, the image-tracking-based geometric correction system 200 may include an acquisition module 210, a tracking module 220, a correspondence determination module 230, and a correction module 240. In some embodiments, the functions corresponding to the image-tracking-based geometric correction system 200 may be executed by the processing device 120.

[0034] The acquisition module 210 may be used to acquire multiple projection maps of a geometric phantom. Each small ball in the multiple projection maps has a correspondence with each small ball in the geometric phantom. The multiple projection maps include a first projection map at a first angle; obtain the correspondence between each first projection small ball and each original small ball in the first projection map. For more content on the acquisition of multiple projection maps of the geometric phantom, reference can be made to Figure 3Step 310 and its related description. For more information on obtaining the correspondence between each first projection sphere and each original sphere in the first projection diagram, reference can be made to Figure 3 Step 320 and its related description.

[0035] The tracking module 220 can be used to track the spheres in the projection diagram based on the similarity between adjacent angular projection diagrams, obtaining a tracking result. For more information on sphere tracking, reference can be made to Figure 3 Step 330 and its related description.

[0036] The correspondence determination module 230 can be used to obtain the correspondence between each sphere in the geometric motif and each sphere in the projection diagram based on the tracking result. For more information on determining the correspondence between spheres, reference can be made to Figure 3 Step 340 and its related description.

[0037] The correction module 240 can be used to perform geometric correction based on the correspondence. For more information on geometric correction, reference can be made to Figure 3 Step 350 and its related description.

[0038] It should be understood that Figure 2 The system and its modules shown can be implemented in various ways. For example, it can be implemented through hardware, software, or a combination of software and hardware. The system and its modules in this specification can be implemented not only by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips and transistors, or programmable hardware devices such as field programmable gate arrays and programmable logic devices, but also by software executed by various types of processors, or by a combination of the above hardware circuits and software (e.g., firmware).

[0039] It should be noted that the above description of the system and its modules is only for convenience of description and as an illustration, and does not limit this specification to the scope of the exemplified embodiments. It can be understood that for those skilled in the art, after understanding the principle of the system, they may, without departing from this principle, make any combination of the modules, or form a subsystem and connect it with other modules.

[0040] Figure 3 is a flowchart of an exemplary image-tracking-based geometric correction method according to some embodiments of this specification. In some embodiments, process 300 can be executed by the processing device 120 or the image-tracking-based geometric correction system 200. For example, process 300 can be stored in a storage device (e.g., storage device 140, the storage unit of the processing device 120) in the form of a program or instruction. When the processor or Figure 2When the module shown executes a program or instruction, process 300 can be implemented. In some embodiments, process 300 can utilize one or more additional operations not described below, and / or can be completed without one or more of the operations discussed below. Additionally, as Figure 3 The order of the operations shown is not restrictive.

[0041] Step 310, obtain multiple projection images of the geometric phantom. Each projection sphere in the multiple projection images has a corresponding relationship with each original sphere in the geometric phantom. The multiple projection images include a first projection image at a first angle. In some embodiments, step 310 can be executed by processing device 120 or acquisition module 210.

[0042] A geometric phantom refers to a phantom used for geometric correction. The geometric phantom can be a cylinder, sphere, prism, cuboid, cube, cone, pyramid, etc. or any combination thereof. In some embodiments, the geometric phantom can contain one or more small spheres (e.g., the small spheres are embedded in the geometric phantom and are stationary relative to the geometric phantom), and the density of the geometric phantom material is different from the density of the small spheres. In some embodiments, the geometric phantom can be transparent or translucent, and the human eye can identify one or more small spheres inside the geometric phantom. In some embodiments, the geometric phantom can be opaque, and it is not easy for the human eye to identify one or more small spheres inside the geometric phantom, but after being scanned by an imaging device (e.g., imaging device 110) and generating an image, one or more small spheres inside the geometric phantom can be identified from the image. In some embodiments, the small spheres are made of high attenuation materials. For example, the small spheres can be made of high attenuation materials such as steel, tin, barium, etc.

[0043] A projection image refers to a two-dimensional image obtained by scanning and / or photographing a geometric phantom. For example, a two-dimensional image obtained by scanning and / or photographing a geometric phantom through an optical imaging device, a medical imaging device (e.g., imaging device 110), etc.

[0044] In some embodiments, the multiple projection images can be obtained from imaging device 110, storage device 140, the storage unit of processing device 120, etc. In some embodiments, acquisition module 210 can obtain the projection images by reading from a storage device, database, calling a data interface, etc.

[0045] In some embodiments, each projection sphere in the multiple projection images has a corresponding relationship with each original sphere in the geometric phantom. For example, as Figure 4 shown, each projection sphere in projection image 420 is the projection of a certain original sphere in geometric phantom 410 in the two-dimensional image.

[0046] In some embodiments, multiple projection images may include a first projection image at a first angle. Each projection sphere in the first projection image is a reference and / or starting projection image for image tracking, and the first angle is a reference and / or starting angle for image tracking. The correspondence between each projection sphere in the first projection image and each original sphere in the geometric phantom is determined. For example, an expert can mark this correspondence. The acquisition module 210 can determine the first projection image and / or the first angle through various methods.

[0047] In some embodiments, the multiple projection images may further include second projection images at other angles continuously acquired. The second projection images are one or more projection images obtained by scanning and / or photographing the geometric phantom when the geometric phantom rotates to one or more angles other than the first angle.

[0048] In some embodiments, the acquisition module 210 can acquire multiple projection images of the geometric phantom through various methods.

[0049] In some embodiments, the acquisition module 210 can capture a preset number of projection images at equal angular intervals during one full rotation of the geometric phantom. The preset number can be determined according to experience, requirements, and / or scanning parameters. Taking the case where the geometric phantom is a cylinder and the preset number is 100 as an example, the acquisition module 210 can capture 100 projection images with an angular interval of 3.6 degrees between adjacent projection images during one full rotation (360 degrees) of the cylinder around its axis.

[0050] In some embodiments, the acquisition module 210 can acquire scanning parameters of the target object. In some embodiments, the scanning parameters are related to the number of images to be captured during the scanning process. For example, if the scanning part is the head, the number of images to be captured is relatively large; if the scanning part is the arm, the number of images to be captured is relatively small. Another example is that the larger the scanning field, the more images to be captured. In some embodiments, the acquisition module 210 can obtain the scanning parameters of the target object from the medical history of the target object, doctor's orders, factory parameters of the scanning device, etc.

[0051] In some embodiments, the acquisition module 210 can determine the angular difference between adjacent angles based on the scanning parameters. For example, if the scanning part is the head and the number of images to be captured is 400, the angular difference between adjacent angles is 0.9 degrees. Another example is that if the scanning part is the arm and the number of images to be captured is 100, the angular difference between adjacent angles is 3.6 degrees.

[0052] Step 320: Obtain the correspondence between each first projection sphere in the first projection image and each original sphere. In some embodiments, step 320 can be executed by the processing device 120 or the acquisition module 210.

[0053] In some embodiments, first, the obtaining module 210 can select a first projection image that meets the angular feature from multiple projection images through an automatic detection algorithm, and use the angle corresponding to the first projection image as the first angle.

[0054] The automatic detection algorithm refers to an algorithm that identifies the first projection image by detecting known features. The known features can include the arrangement features of the projection balls in the projection image, etc.

[0055] In some embodiments, the automatic detection algorithm can use the projection image in multiple projection images whose arrangement feature of the projection balls is the same as or closest to the preset arrangement feature as the first projection image. The preset arrangement feature can be set in advance according to experience or requirements. For example, the arrangement feature of all projection balls when a certain projection ball is located on the center line of the projection image can be used as the preset arrangement feature.

[0056] In some embodiments, the geometric phantom further includes a large ball, and the known features can also include the projection features of the large ball (for example, the center coordinates of the large ball). The automatic detection algorithm can use the projection image in multiple projection images whose projection features of the large ball are the same as or closest to the preset projection features of the large ball as the first projection image. The preset projection features of the large ball can be set in advance according to experience or requirements. For example, the preset projection features of the large ball can be that the projection coordinates of the center of the large ball in the projection image are (17, 19).

[0057] In some embodiments, the projection coordinates are two-dimensional coordinates based on pixels, with the origin at the upper left corner of the projection image (the first projection image and / or the second projection image), and the unit is pixels.

[0058] In the prior art, generally, the geometric phantom is manually placed at the first angle first, and then the first projection image is taken, which has a very high requirement for the accuracy of manual placement. In the embodiments of the present application, by first taking multiple projection images and then automatically identifying the first projection image and / or the first angle from the multiple projection images, manual participation can be reduced, and the accuracy and efficiency can be improved.

[0059] Secondly, the obtaining module 210 can automatically identify the correspondence between each first small ball in the first projection image and each original small ball in the geometric phantom. In some embodiments, this correspondence can be obtained from the storage device 140, the storage unit of the processing device 120, etc. In some embodiments, the obtaining module 210 can obtain this correspondence by reading from the storage device, database, calling the data interface, etc.

[0060] In some embodiments, the obtaining module 210 can obtain the correspondence between each first projection small ball and each original small ball in the first projection image by other methods. For example, by the method of expert calibration. Another example is by the method of training a machine learning model.

[0061] Step 330: Track the projection balls based on the similarity between the projection diagrams of adjacent angles to obtain a tracking result. In some embodiments, step 330 may be executed by the processing device 120 or the tracking module 220.

[0062] The similarity between the projection diagrams of adjacent angles refers to the property that the projection diagrams of adjacent angles are similar or close. For example, the central coordinates of the projection balls in the projection diagrams of adjacent angles are relatively close, and the arrangements of the projection balls in the projection diagrams of adjacent angles are similar, etc.

[0063] The tracking result refers to the result reflecting the changes of the projection balls between the projection diagrams. For example, the tracking result may include the position changes, movement trajectories, identifiers, etc. of the projection balls.

[0064] In some embodiments, the tracking module 220 may track the projection balls in various ways to obtain a tracking result.

[0065] In some embodiments, the tracking module 220 may track the projection balls based on the projection coordinates of the projection balls. In some embodiments, the tracking module 220 may obtain the projection coordinates of each first projection ball and the coordinates of each second projection ball in the second projection diagram. For each first projection ball, when the absolute value of the difference between the projection coordinates of a certain second projection ball in the second projection diagram and the first projection ball is the smallest, it is determined that the second projection ball and the first projection ball are the projections of the same original ball.

[0066] As Figure 5 shown, in some embodiments, the tracking module 220 may obtain the projection coordinates of each first projection ball (A1, A2, A3, A4... A N ) in the first projection diagram 510 and the coordinates of each second projection ball (B1, B2, B3, B4... B N ) in the second projection diagram 520. Taking A1 as an example, if the absolute value of the difference between the projection coordinates of a certain second projection ball (for example, B2) in B1, B2, B3, B4... B N and A1 is the smallest, it is determined that B2 and A1 are the projections of the same original ball.

[0067] In some embodiments, the tracking module 220 may, through the above method, track the second projection balls in all the second projection diagrams based on the projection diagrams of adjacent angles to obtain a tracking result. For example, based on the first projection diagram (the first projection diagram) and the second projection diagram of an adjacent angle, track the second projection balls in the second projection diagram; based on the second projection diagram and the third projection diagram of an adjacent angle, track the second projection balls in the third projection diagram;...; based on the 99th projection diagram and the 100th projection diagram of an adjacent angle, track the second projection balls in the 100th projection diagram.

[0068] In some embodiments, the tracking module 220 may use an identification model to track the small balls in the projection images. For the identification model and its training, refer to Figure 6 and its description.

[0069] Step 340, obtaining the correspondence between each original small ball and each projected small ball based on the tracking result. In some embodiments, step 340 may be executed by the processing device 120 or the correspondence determination module 230.

[0070] The correspondence between each original small ball and each projected small ball refers to the correspondence between the voxel coordinates and the projection coordinates of each small ball in the geometric voxel and the projection images at various angles. Determining this correspondence is a very important step in the calculation of geometric correction.

[0071] In some embodiments, the voxel coordinates are three-dimensional coordinates in space, with the origin at the center of the voxel and the unit being a specific measurement unit, such as millimeters. In some embodiments, the projection coordinates are two-dimensional coordinates based on pixels, with the origin at the upper left corner of the projection image (the first projection image and / or the second projection image) and the unit being pixels.

[0072] In some embodiments, the relationship determination module 230 may determine this correspondence based on the tracking result. For example, based on the original small ball and the projected small balls of the original small ball obtained by tracking in each projection image, determine the relationship between the voxel coordinates of the original small ball and the projection coordinates of the projected small balls in each projection image.

[0073] Step 350, performing geometric correction based on the correspondence. In some embodiments, step 350 may be executed by the processing device 120 or the correction module 240.

[0074] Geometric correction refers to the process of eliminating or correcting geometric errors. Through geometric correction, geometric artifacts in the reconstructed image caused by geometric errors can be eliminated or reduced.

[0075] In some embodiments, the correction module 240 may perform geometric correction based on the correspondence by methods such as two-metal-small-ball geometric correction and iterative geometric correction.

[0076] It should be noted that the above description of the process 300 is only for illustration and explanation, and does not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to the process 300 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

[0077] Figure 6 is a schematic diagram of the identification model and its training shown in some embodiments of this specification.

[0078] An identification model refers to a model used for the corresponding relationship of the same sphere in the projection maps of adjacent angles. In some embodiments, the identification model can be a machine learning model, and the machine learning model can include, but is not limited to, one or a combination of multiple types such as neural network models, support vector machine models, k-nearest neighbor models, decision tree models, etc. Among them, the neural network model can include one or a combination of multiple types such as CNN, LeNet, GoogLeNeT, ImageNet, AlexNet, VGG, ResNet, etc.

[0079] As Figure 6 shown, in some embodiments, the input of the identification model includes the projection maps of adjacent angles; the output of the identification model includes the corresponding relationship of the same sphere in the projection maps of adjacent angles.

[0080] As Figure 6 shown, in some embodiments, an initial identification model 610 can be trained based on a large number of labeled training samples to update the parameters of the initial identification model to obtain a trained identification model 620.

[0081] In some embodiments, the processing device can obtain multiple training samples, and each training sample includes the sample projection maps of adjacent angles and the corresponding relationship of the same sphere in the sample projection maps of adjacent angles. In some embodiments, each training sample can include the sample projection maps of adjacent angles as the input of the training model, and the corresponding relationship of the same sphere in the sample projection maps as the label.

[0082] In some embodiments, the corresponding relationship of the same sphere in the sample projection maps corresponding to the label can be obtained by performing manual or automatic processing on the sample projection maps of adjacent angles. The label can be added manually or automatically, or added by other means, and this embodiment does not limit this.

[0083] In some embodiments, the processing device can obtain multiple training samples, including their corresponding labels, by reading from a database, storage device, or calling a data interface.

[0084] In some embodiments, the processing device can process the sample projection maps of adjacent angles through the identification model to obtain the predicted corresponding relationship of the same sphere in the sample projection maps.

[0085] In some embodiments, the processing device can construct a loss function based on the predicted corresponding relationship and the label of the training sample (the corresponding relationship of the same sphere in the sample projection maps), and update the initial identification model according to the loss function to obtain a trained identification model.

[0086] The loss function can reflect the magnitude of the difference between the predicted correspondence and the label. The processing device can adjust the parameters of the recognition model based on the loss function to reduce the difference between the predicted correspondence and the label. For example, by continuously adjusting the parameters of the recognition model, the value of the loss function is reduced or minimized.

[0087] In some embodiments, the recognition model can also be obtained according to other training methods. For example, an appropriate initial learning rate (e.g., 0.1) and a learning rate decay strategy are set for the recognition model, and the recognition model is obtained through training based on the labeled training samples. This application does not limit this here.

[0088] It should be noted that the above description of process 600 is only for illustration and explanation, and does not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to process 600 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification. In some embodiments, Figure 6 the model generation process described in Figure 3 and the geometric correction process (e.g., Figure 6 tracking the small ball described in

[0089] can be executed on different processing devices. For example,

[0090] the model generation process described in

[0089] can be performed on the processing device of the imaging device manufacturer, while part or all of the geometric correction process can be performed on the processing device of the user of the imaging device (such as a hospital).

[0089] In some embodiments of this specification, (1) the relationship between adjacent shooting angle projections is used to track the small ball, and the tracking is more accurate; (2) the reference and / or starting projection map is automatically selected, reducing or avoiding manual participation, and improving the accuracy and efficiency of the correction work.

[0090] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only an example and does not constitute a limitation to this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are proposed in this specification, so such modifications, improvements, and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

[0091] Meanwhile, this specification uses specific terms to describe the embodiments of this specification. For example, "an embodiment", "one embodiment", and / or "some embodiments" mean a certain feature, structure, or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "an embodiment" or "one embodiment" or "an alternative embodiment" mentioned twice or more at different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0092] Moreover, unless clearly stated in the claims, the order of the processing elements and sequences, the use of numerical and alphabetical characters, or the use of other names in this specification are not used to limit the order of the processes and methods in this specification. Although some currently considered useful embodiments of the invention are discussed through various examples in the above disclosure, it should be understood that such details only serve the purpose of illustration. The appended claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that conform to the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only through software solutions, such as installing the described system on existing servers or mobile devices.

[0093] Similarly, it should be noted that, in order to simplify the expression of the disclosure in this specification and thus help the understanding of one or more embodiments of the invention, in the previous description of the embodiments of this specification, sometimes multiple features are merged into one embodiment, drawing, or description thereof. However, this disclosure method does not mean that the features required by the subject of this specification are more than those mentioned in the claims. In fact, the features of the embodiments are fewer than all the features of the individual embodiments disclosed above.

[0094] In some embodiments, numbers are used to describe the components and the quantity of attributes. It should be understood that such numbers used for the description of embodiments are, in some examples, modified by the modifiers "about", "approximate", or "substantially". Unless otherwise stated, "about", "approximate", or "substantially" indicate that the said numbers allow a variation of ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, and such approximate values can change according to the characteristics required by individual embodiments. In some embodiments, the numerical parameters should consider the specified significant digits and adopt the method of retaining the general number of digits. Although the numerical ranges and parameters used in some embodiments of this specification to confirm the breadth of their scope are approximate values, in specific embodiments, such numerical settings are made as precise as possible within the feasible range.

[0095] For each patent, patent application, patent application publication, and other materials cited in this specification, such as articles, books, specifications, publications, documents, etc., their entire contents are hereby incorporated into this specification by reference. This excludes the application history files that are inconsistent with or conflict with the content of this specification, as well as the files that limit the broadest scope of the claims of this specification (currently or subsequently appended to this specification). It should be noted that if there are any inconsistencies or conflicts between the descriptions, definitions, and / or uses of terms in the supplementary materials of this specification and the content described in this specification, the descriptions, definitions, and / or uses of terms in this specification shall prevail.

[0096] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A geometric correction method based on image tracking, characterized in that The method includes: Obtaining multiple projection images of a geometric phantom, where each projection small ball in the multiple projection images has a corresponding relationship with each original small ball in the geometric phantom, and the multiple projection images include a first projection image at a first angle and second projection images at other continuously obtained angles; Selecting, through an automatic detection algorithm, the first projection image that meets the angle feature from the multiple projection images; Taking the angle corresponding to the first projection image as the first angle, and the first angle is the reference and / or starting angle for image tracking; Automatically identifying or calibrating by an expert the corresponding relationship between each first projection small ball and each original small ball in the first projection image; Tracking the projection small balls based on the similarity between projection images at adjacent angles to obtain a tracking result, where the tracking result refers to a result reflecting the change of the projection small balls between the projection images; Obtaining the corresponding relationship between each original small ball and each projection small ball based on the tracking result, and the corresponding relationship between each original small ball and each projection small ball refers to the corresponding relationship between the phantom coordinates of each original small ball in the geometric phantom and the projection coordinates of each projection small ball in the projection images at each angle; Performing geometric correction based on the corresponding relationship between each original small ball and each projection small ball; The tracking the projection small balls based on the similarity between projection images at adjacent angles to obtain a tracking result includes: Tracking the second projection small balls in all the second projection images based on the projection images at adjacent angles to obtain a tracking result.

2. The method according to claim 1, wherein The obtaining multiple projection images of a geometric phantom includes: Obtaining the scanning parameters of the target object, where the scanning parameters are related to the number of images to be captured; Determining the angle difference between adjacent angles based on the scanning parameters.

3. The method according to claim 1, wherein The tracking the projection small balls based on the similarity between projection images at adjacent angles includes: For each first projection small ball, When the absolute value of the difference in projection coordinates between a certain second projection small ball in the second projection image at an angle adjacent to the first projection image and the first projection small ball is the smallest, it is determined that the second projection small ball and the first projection small ball are projections of the same original small ball.

4. The method according to claim 1, characterized in that The tracking the projection small balls based on the similarity between the projection images at adjacent angles includes: Using an identification model to track the projection small balls.

5. The method according to claim 4, wherein: The input of the identification model includes projection images at adjacent angles; The output of the identification model includes the corresponding relationship of the same projection small ball in the projection images at adjacent angles.

6. A geometric correction system based on image tracking, characterized in that, The system includes: An acquisition module, configured to acquire multiple projection images of a geometric phantom, where each projection small ball in the multiple projection images has a corresponding relationship with each original small ball in the geometric phantom, and the multiple projection images include a first projection image at a first angle and second projection images at other angles continuously acquired; select the first projection image that meets the angle feature from the multiple projection images through an automatic detection algorithm; use the angle corresponding to the first projection image as the first angle, and the first angle is the reference and / or starting angle for image tracking; automatically identify or have an expert calibrate the corresponding relationship between each first projection small ball and each original small ball in the first projection image; A tracking module, configured to track the projection small balls based on the similarity between the projection images at adjacent angles to obtain a tracking result, where the tracking result refers to a result reflecting the change of the projection small balls between the projection images, including: based on the projection images at adjacent angles, tracking the second projection small balls in all the second projection images to obtain a tracking result; A corresponding relationship determination module, configured to obtain the corresponding relationship between each original small ball and each projection small ball based on the tracking result, and the corresponding relationship between each original small ball and each projection small ball refers to the corresponding relationship between the phantom coordinates of each original small ball in the geometric phantom and the projection coordinates of each projection small ball in the projection images at each angle; A correction module, configured to perform geometric correction based on the corresponding relationship between each original small ball and each projection small ball.

7. A geometric correction device based on image tracking, characterized in that The apparatus includes: A display device, configured to display at least one of the projection images of the geometric phantom and the corresponding relationship between each small ball in the projection images; At least one storage medium, storing computer instructions; and At least one processor, configured to execute the computer instructions to implement the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, where the storage medium stores computer instructions, and when a computer reads the computer instructions, the computer executes the method according to any one of claims 1 to 5.

Images