Automatic section positioning method and device in 3D heart CT, medium and product

Abstract

The invention discloses an automatic section positioning method and device in 3D heart CT, a medium and a product, and relates to the field of image section positioning. The method comprises the following steps: performing image resampling on a 3D heart CT image aligned to a standard 3D heart atlas space by using a block random sampling strategy to obtain training sample data; training the positioning regression network by adopting the training sample data in a self-supervision mode; inputting the query image into the trained positioning regression network to obtain a tangent plane position of the query image in the map space; obtaining a candidate position of the query image in the target 3D heart CT space through rigid transformation based on the position of the tangent plane; performing image resampling at the candidate position to obtain a candidate database; and a candidate image section with the highest similarity is found in the candidate database by using an image similarity algorithm, and the position of the query image in the target 3D heart CT space is obtained, so that positioning and viewing of any section are realized, and the positioning efficiency of any section is improved.

Description

Technical Field

[0001] The present application relates to the field of image section positioning, and in particular to a method, device, equipment, medium and product for automatic section positioning in 3D cardiac CT. Background Art

[0002] In actual clinical practice, the cross-sectional views of cardiac CT can help doctors quickly locate and view images, and provide data reference for doctors to diagnose diseases, plan cross-sectional views, and plan surgery. However, most of the standard cross-sectional views of cardiac CT are based on specific cardiac anatomical structure definitions, and due to large individual differences, cross-sectional positioning is heavily dependent on expert experience; and because the search space of 3D cardiac CT is large, manual cross-sectional positioning requires doctors to spend a lot of time and is inefficient.

[0003] The existing methods for section positioning of cardiac CT can be divided into two categories: one is the traditional method, which mainly locates the target section of the heart by manually designing and selecting features, but this method is sensitive to the selection of features, and may result in serious positioning errors and poor generalization. In addition, due to the large individual differences, section positioning is heavily dependent on expert experience; and because the search space of 3D cardiac CT is large, manual section positioning takes a lot of time for doctors and is inefficient. The second is a method based on deep learning, which mainly obtains prior knowledge of the entire 3D cardiac CT image (for example: segmentation of anatomical structures, marking key points, etc.) by training a neural network model to perform section positioning. During the implementation process, a large number of high-quality anatomical structure samples need to be manually labeled, which has the disadvantages of being time-consuming and labor-intensive.

[0004] In addition, existing technologies are methods and models designed for specific sections and cannot locate and view any section. However, in actual clinical cardiac CT examinations and surgical planning, the sections are not fixed, but different sections are selected based on the differences in different individuals and different diseases for diagnosis and planning.

[0005] In summary, most of the existing slice positioning methods are based on prior knowledge such as segmentation or key points, and are heavily dependent on time-consuming and labor-intensive manual labeling or expert experience. In addition, the trained models are limited to the positioning of specific slices and cannot achieve the positioning and viewing of arbitrary slices. Summary of the invention

[0006] The purpose of this application is to provide a method, device, equipment, medium and product for automatic section positioning in 3D cardiac CT, which can realize the positioning and viewing of any section and at the same time improve the positioning efficiency of any section.

[0007] To achieve the above objectives, this application provides the following solutions:

[0008] In a first aspect, the present application provides a method for automatic section positioning in 3D cardiac CT, comprising:

[0009] Generate a standard 3D cardiac atlas based on 3D cardiac CT images aligned into a unified space;

[0010] Using a block random sampling strategy, the CT images in the standard 3D cardiac atlas space are resampled to obtain 2D sections, the positions of the 2D sections are determined, and training sample data are generated;

[0011] Using the training sample data to train the positioning regression network in a self-supervised manner until the set conditions are met, thereby obtaining a trained positioning regression network;

[0012] Input the query image into the trained positioning regression network to obtain the section position of the query image in the atlas space;

[0013] Based on the section position of the query image, a candidate position of the query image in the target 3D cardiac CT space is obtained by rigid transformation of a standard 3D cardiac atlas and a target 3D cardiac CT space;

[0014] Performing image resampling at candidate positions in the target 3D cardiac CT space to obtain candidate image slices to generate a candidate database;

[0015] By using an image similarity algorithm, a candidate image section with the highest similarity to the query image is found in the candidate database, and the position of this candidate image section is used as the positioning result of the query image to complete the automatic positioning of the section in 3D cardiac CT.

[0016] Optionally, generating a standard 3D cardiac atlas based on the 3D cardiac CT images aligned to a unified space includes:

[0017] Randomly select a CT space from all 3D cardiac CT spaces as a CT image reference space, and align all remaining 3D cardiac CT images to the CT image reference space by using a symmetric differential homeomorphic deformable nonlinear registration algorithm;

[0018] The standard 3D cardiac atlas is generated based on all CT images aligned in the CT image reference space using an averaging algorithm.

[0019] Optionally, generating the standard 3D cardiac atlas based on all CT images aligned to the CT image reference space using an averaging algorithm specifically includes:

[0020] Performing an average process on all CT images aligned to the CT image reference space to obtain an initial template;

[0021] Using the symmetric diffeomorphic deformable nonlinear registration algorithm, registering all CT images aligned to the CT image reference space to the initial template;

[0022] The initial template is adjusted using a sharpening filter to obtain the standard 3D heart atlas.

[0023] Optionally, the loss function of the positioning regression network is:

[0024]

[0025] Where λ represents the weight of the translation loss, r represents the rotation in the true rigid transformation, and t represents the translation in the true rigid transformation. represents the rotation in the predicted rigid transformation, represents the translation in the predicted rigid transformation, represents the rotation in the predicted rigid transformation and the geodesic distance error between the rotation r in the true rigid transformation, represents the translation in the predicted rigid transformation The mean square error between the translation t in the true rigid transformation, Loss total Indicates the loss value.

[0026] Optionally, the candidate position of the query image in the target 3D cardiac CT space is expressed as:

[0027]

[0028] In the formula, represents the candidate position, RT represents the rigid transformation between the standard 3D cardiac atlas and the target 3D cardiac CT space, Represents the slice position of the query image in the atlas space.

[0029] Optionally, the candidate image slice with the highest similarity to the query image is expressed as:

[0030]

[0031] In the formula, represents the candidate image slice with the highest similarity to the query image, represents the i-th candidate image slice in the candidate database, m represents the number of candidate image slices, Represents the i-th candidate image slice and the image similarity of the query image q.

[0032] In a second aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the automatic section positioning method in 3D cardiac CT described in any one of the above.

[0033] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the above-described methods for automatic section positioning in 3D cardiac CT.

[0034] In a fourth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any one of the above-mentioned methods for automatic section positioning in 3D cardiac CT.

[0035] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0036] The present application provides a method, device, medium and product for automatic section positioning in 3D cardiac CT. By generating a standard 3D cardiac atlas, the problem of inconsistent standard section views caused by large individual differences in cardiac structure is solved. Guided by the standard 3D cardiac atlas, the automatic positioning and viewing of any section in 3D cardiac CT is achieved in a coarse-to-fine manner (i.e., the final query image position positioning is achieved by the candidate position). Moreover, in a self-supervised manner, the positioning regression network is trained using the training sample data, eliminating the requirement for manual labeling, thereby improving the positioning efficiency of any section. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 This is a diagram of an application environment of an automatic section positioning method in 3D cardiac CT in one embodiment of the present application;

[0039] Figure 2 A schematic diagram of a flow chart of an automatic section positioning method in 3D cardiac CT provided in one embodiment of the present application;

[0040] Figure 3 This is a diagram of an implementation architecture of an automatic section positioning method in 3D cardiac CT provided in one embodiment of the present application;

[0041] Figure 4 An array diagram of automatic section positioning results provided by an embodiment of the present application;

[0042] Figure 5 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0044] In order to make the purpose, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0045] The automatic section positioning method in 3D cardiac CT provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the query image to the server 104. After the server 104 receives the query image, for the query image, the server 104 inputs the query image into the trained positioning regression network to obtain the section position of the query image in the atlas space. Based on the section position of the query image, the candidate position of the query image in the target 3D cardiac CT space is obtained by rigid transformation of the standard 3D cardiac atlas and the target 3D cardiac CT space. Image resampling is performed at the candidate position in the target 3D cardiac CT space to obtain the candidate image section to generate a candidate database. Using the image similarity algorithm, the candidate image section with the highest similarity to the query image is found in the candidate database, and the position of this candidate image section is used as the positioning result of the query image to complete the automatic positioning of the section in the 3D cardiac CT. The server 104 can feedback the obtained positioning result of the query image to the terminal 102. In addition, in some embodiments, the automatic section positioning method in 3D cardiac CT can also be implemented independently by the server 104 or the terminal 102. For example, the terminal 102 can directly perform automatic positioning on the query image, or the server 104 can obtain the query image from the data storage system and perform automatic positioning processing on the query image.

[0046] The terminal 102 may be, but is not limited to, various desktop computers, laptop computers, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart vehicle-mounted devices, etc. The portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers, or may be a cloud server.

[0047] In an exemplary embodiment, Figure 2 As shown, a method for automatic section positioning in 3D cardiac CT is provided. The method is executed by a computer device, and specifically can be executed by a computer device such as a terminal or a server alone, or can be executed by a terminal and a server together. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in the example is used as an example to illustrate the method, which includes the following steps 200 to 206. Among them:

[0048] Step 200: Generate a standard 3D cardiac atlas based on the 3D cardiac CT images aligned into a unified space.

[0049] Step 201: Using a block random sampling strategy, resample the CT image in the standard 3D cardiac atlas space to obtain a 2D section, determine the position of the 2D section, and generate training sample data.

[0050] Step 202: Using training sample data to train the positioning regression network in a self-supervised manner until the set conditions are met, thereby obtaining a trained positioning regression network.

[0051] Step 203: Input the query image into the trained positioning regression network to obtain the section position of the query image in the atlas space.

[0052] Step 204: Based on the section position of the query image, a candidate position of the query image in the target 3D cardiac CT space is obtained by rigid transformation of the standard 3D cardiac atlas and the target 3D cardiac CT space.

[0053] Step 205: performing image resampling at the candidate position in the target 3D cardiac CT space to obtain candidate image slices to generate a candidate database.

[0054] Step 206: Using an image similarity algorithm, find the candidate image slice with the highest similarity to the query image in the candidate database, and use the position of this candidate image slice as the positioning result of the query image to complete the automatic positioning of the slice in the 3D cardiac CT.

[0055] By implementing the above steps 201 to 206, the present application can realize the positioning and viewing of any section. In addition, the present application uses training sample data to train the positioning regression network in a self-supervised manner, which can eliminate the requirement of manual labeling and thus improve the positioning efficiency of any section.

[0056] In another exemplary embodiment of the present application, in order to solve the problem of inconsistent standard section views due to large individual differences in cardiac structure, the accuracy of slice positioning is improved by generating a standard 3D cardiac atlas by unifying the coordinate system and registering it to the 3D cardiac CT space under the unified coordinates. Based on this, the above step 200 is replaced by the following steps 301 to 302:

[0057] Step 301: randomly select a CT space from all 3D cardiac CT spaces as a CT image reference space, and align all remaining 3D cardiac CT images to the CT image reference space by using a symmetric diffeomorphic deformable nonlinear registration algorithm.

[0058] Step 302: Generate a standard 3D cardiac atlas based on all CT images aligned to the CT image reference space using an averaging algorithm.

[0059] The specific implementation process of step 302 can be described as follows:

[0060] (1) All CT images aligned to the CT image reference space are averaged to obtain the initial template.

[0061] (2) A symmetric diffeomorphic deformable nonlinear registration algorithm is used to register all CT images aligned in the CT image reference space to the initial template.

[0062] (3) Use a sharpening filter to adjust the initial template to obtain a standard 3D heart atlas.

[0063] In actual application, after obtaining the initial template in step (2), all deformed CT images can be averaged to create a new template, and then all transformations from step (2) are averaged to create a single transformation. The single transformation is applied to distort the atlas toward the true average shape, and then the initial module is continuously adjusted using a sharpening filter to enhance the edges.

[0064] In another exemplary embodiment of the present application, in order to train a positioning regression network in a unified coordinate system to eliminate the requirement for manual annotation and realize automatic section location based on any input section instead of a fixed section, the present application uses a self-supervised method to train customized atlas prompts for different input query images to locate the target section. Based on this, the above step 202 is replaced by the following steps 401 to 402:

[0065] Step 401: Parameterize the position of the 2D slice in the 3D cardiac CT space, denoted as p i = {r i ,t i}, that is, a rigid transformation represented by rotation and translation.

[0066] Step 402: To obtain n pairs of training data (2D sections and their corresponding positions), the present application utilizes a block random sampling strategy to resample a large number of 2D sections in a 3D cardiac CT in a unified coordinate system.

[0067] Step 403: Use the large amount of sampled paired data to train the positioning regression network PosNet, which is denoted by γ(·; μ), where μ is a parameter in the positioning regression network PosNet. The positioning regression network PosNet consists of two parts: the encoder module γ enc (·;μ enc ) and the positioning prediction module γ pos (·;μ pos ). enc is the encoder module γ enc The parameters in μ pos is the positioning prediction module γ pos Each 2D slice in the training data is input to the encoder module γ enc (·;μ enc 0 to generate a feature vector f i ,for:

[0068]

[0069] Step 404: By transforming the feature vector Input to the positioning prediction module γ pos (·;μ pos ) to predict the standard 3D cardiac atlas The position of the cut surface in for:

[0070]

[0071] Step 405: The loss function is designed as the sum of the mean square error loss between the geodesic distance and translation of the 3D rotation in the rigid transformation and the true distance and translation, denoted as:

[0072]

[0073] In the formula, λ represents the weight of the translation loss, r represents the rotation in the true rigid transformation, and t represents the translation in the true rigid transformation. represents the rotation in the predicted rigid transformation, represents the translation in the predicted rigid transformation, represents the rotation in the predicted rigid transformation and the geodesic distance error between the rotation r in the true rigid transformation, represents the translation in the predicted rigid transformation The mean square error between the translation t in the true rigid transformation, Loss total Indicates the loss value.

[0074] In another exemplary embodiment of the present application, in order to achieve coarse positioning of the query image, step 204 provided above in the present application can be replaced by steps 501 to 504:

[0075] Step 501: The query image q is input into the trained positioning regression network PosNet to obtain the query image q in the standard 3D heart atlas. The position in

[0076]

[0077] Step 502: Standard 3D cardiac atlas And the query image q in the standard 3D heart atlas Position in Graph hints that make up the query image q

[0078] Step 503: For rough positioning, standard 3D cardiac atlas It is rigidly registered to the target 3D cardiac CT space v to obtain a rigid transformation RT between the two.

[0079] Step 504: The rigid transformation RT is applied to the position Thus, the rough position (i.e., candidate position) of the target 3D cardiac CT space v is obtained, which is recorded as

[0080] In another exemplary embodiment of the present application, after the rough positioning, in order to achieve the precise positioning of the query image, so as to achieve the automatic positioning of any section in the 3D cardiac CT from a rough to a fine manner under the guidance of the atlas prompt, the above steps 205 and 206 provided in the present application can be replaced by steps 601 to 602:

[0081] Step 601: Resample candidate image slices at a coarse position (ie, candidate position) in the target 3D cardiac CT space v to form a candidate database.

[0082] Step 602: Use image similarity to query the most similar slice And further get the position of the target 3D heart CT slice. The image similarity method used here is structural similarity, which is the representation of structural similarity by the human visual system and evaluates the similarity of images from aspects such as brightness, contrast, and structure.

[0083] Among them, the candidate image slice with the highest similarity to the query image is expressed as:

[0084]

[0085] In the formula, represents the candidate image slice with the highest similarity to the query image, represents the i-th candidate image slice in the candidate database, m represents the number of candidate image slices, Represents the i-th candidate image slice and the image similarity of the query image q.

[0086] In summary, under the guidance of the generated atlas hint corresponding to the query image q, the standard 3D heart atlas and target 3D cardiac CT space The rigid transformation of the query image in the target 3D cardiac CT space is obtained by Candidate location Then, the final target section is determined by using the image similarity through the candidate sections generated at the candidate positions. and its corresponding position

[0087] Based on the above description, the automatic section positioning method in 3D cardiac CT provided by the present application, and its overall implementation architecture can be found in Figure 3 , which can be summarized into two parts: atlas prompt and section positioning. The atlas prompt part includes generating a standard 3D cardiac atlas and training a positioning regression network in a self-supervised manner, i.e., steps 200 to 202. The section positioning part includes coarse positioning and fine positioning, i.e., steps 203 to 206.

[0088] Furthermore, experimental means are used to illustrate the advantages of the automatic section positioning method in the 3D cardiac CT provided in the present application.

[0089] Figure 4 The first column in is the three query slices in arbitrary CT space: 2C (long axis 2-chamber center), 4C (long axis 4-chamber center) and X (arbitrary slice). Figure 4 For Case 1 and Case 2, the first column is the positioning result of this application, the second column is the annotation result of the radiologist, and the third column is the visual comparison result of this application and the radiologist's result. Figure 4 In the figure, the green one is the positioning result of this application, and the rose-red one is the result marked by the radiology expert. Figure 4 Case 1 and Case 2 in are query results corresponding to three query sections in different target CT spaces. In detail, Case 1 is the result in the first target 3D cardiac CT space, the first sub-image in the first row is the section corresponding to 2C located by the query of this application, the third sub-image in the first row is the section corresponding to 2C marked by the doctor, and the fourth sub-image in the first row is a visualization of the section located by the query of this application and the section marked by the doctor. The second sub-image in the second row is the section corresponding to 4C located by the query of this application, the third sub-image in the second row is the section corresponding to 4C marked by the doctor, and the fourth sub-image in the second row is a visualization of the section located by the query of this application and the section marked by the doctor. The second sub-image in the third row is the section corresponding to X located by the query of this application, the third sub-image in the third row is the section corresponding to X marked by the doctor, and the fourth sub-image in the third row is a visualization of the section located by the query of this application and the section marked by the doctor. Figure 4 Case 2 in FIG. 1 is the result in the second target 3D cardiac CT space, and its detailed description is similar to that of Case 1, so it will not be repeated here.

[0090] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store 3D cardiac CT section positioning data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for automatic section positioning in 3D cardiac CT is implemented.

[0091] Those skilled in the art will understand that Figure 5The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0092] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0093] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0094] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0095] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0096] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0097] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0098] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for automatic section positioning in 3D cardiac CT, characterized in that: The automatic section positioning method in 3D cardiac CT comprises: Generate a standard 3D cardiac atlas based on 3D cardiac CT images aligned into a unified space; Using a block random sampling strategy, the CT images in the standard 3D cardiac atlas space are resampled to obtain 2D sections, the positions of the 2D sections are determined, and training sample data are generated; Using the training sample data to train the positioning regression network in a self-supervised manner until the set conditions are met, thereby obtaining a trained positioning regression network; Input the query image into the trained positioning regression network to obtain the section position of the query image in the atlas space; Based on the section position of the query image, a candidate position of the query image in the target 3D cardiac CT space is obtained by rigid transformation of a standard 3D cardiac atlas and a target 3D cardiac CT space; Performing image resampling at candidate positions in the target 3D cardiac CT space to obtain candidate image slices to generate a candidate database; By using an image similarity algorithm, a candidate image section with the highest similarity to the query image is found in the candidate database, and the position of this candidate image section is used as the positioning result of the query image to complete the automatic positioning of the section in 3D cardiac CT.

2. The automatic section positioning method in 3D cardiac CT according to claim 1, characterized in that: Generate a standard 3D cardiac atlas based on 3D cardiac CT images aligned to a unified space, including: Randomly select a CT space from all 3D cardiac CT spaces as a CT image reference space, and align all remaining 3D cardiac CT images to the CT image reference space by using a symmetric differential homeomorphic deformable nonlinear registration algorithm; The standard 3D cardiac atlas is generated based on all CT images aligned in the CT image reference space using an averaging algorithm.

3. The automatic section positioning method in 3D cardiac CT according to claim 2, characterized in that: Generating the standard 3D cardiac atlas based on all CT images aligned to the CT image reference space using an averaging algorithm specifically includes: Performing an average process on all CT images aligned to the CT image reference space to obtain an initial template; Using the symmetric diffeomorphic deformable nonlinear registration algorithm, registering all CT images aligned to the CT image reference space to the initial template; The initial template is adjusted using a sharpening filter to obtain the standard 3D heart atlas.

4. The automatic section positioning method in 3D cardiac CT according to claim 1, characterized in that: The loss function of the positioning regression network is: Where λ represents the weight of the translation loss, r represents the rotation in the true rigid transformation, and t represents the translation in the true rigid transformation. represents the rotation in the predicted rigid transformation, represents the translation in the predicted rigid transformation, represents the rotation in the predicted rigid transformation and the geodesic distance error between the rotation r in the true rigid transformation, represents the translation in the predicted rigid transformation The mean square error between the translation t in the true rigid transformation, Loss total Indicates the loss value.

5. The automatic section positioning method in 3D cardiac CT according to claim 1, characterized in that: The candidate position of the query image in the target 3D cardiac CT space is expressed as: In the formula, represents the candidate position, RT represents the rigid transformation between the standard 3D cardiac atlas and the target 3D cardiac CT space, Represents the slice position of the query image in the atlas space.

6. The automatic section positioning method in 3D cardiac CT according to claim 1, characterized in that: The candidate image slice with the highest similarity to the query image is expressed as: In the formula, represents the candidate image slice with the highest similarity to the query image, represents the i-th candidate image slice in the candidate database, m represents the number of candidate image slices, Represents the i-th candidate image slice and the image similarity of the query image q.

7. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the automatic section positioning method in 3D cardiac CT according to any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for automatic section positioning in 3D cardiac CT according to any one of claims 1 to 6 is implemented.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for automatic section positioning in 3D cardiac CT according to any one of claims 1 to 6 is implemented.

Images