3D fetal heart standard section identification method and system, electronic equipment and storage medium

Abstract

The invention provides a 3D fetal heart standard section recognition method and system, electronic equipment and a storage medium, and the method comprises the steps: generating a Gaussian noise image based on Gaussian distribution, inputting the Gaussian noise image and a 3D fetal heart image into an image condition potential U-Net network model, and carrying out the feature fusion; inputting the fusion features into a variational auto-encoder (VAE) to generate a 2D standard section of the 3D fetal heart image; and inputting the 2D standard section into the regression model to obtain a 2D standard section of the 3D fetal heart image. The Gaussian noise image is generated on the basis of Gaussian distribution, the photographed 3D fetal heart image is subjected to dysphoria processing, the 2D standard section is generated, the generated 2D standard section is generated on the basis of the 3D fetal heart image and has certain correlation with the 3D fetal heart image, and the 2D standard section generated in the mode serves as a reference section. And the finally obtained 2D fetal heart tangent plane is more standard.

Description

Technical Field

[0001] The present invention relates to the field of image recognition, and more specifically, to a 3D fetal heart standard section recognition method, system, electronic device and storage medium. Background Art

[0002] Currently, the main steps of the 3D standard section recognition method are: randomly generating a 2D-like standard section, and then regressing the 2D-like standard section into a 2D standard section through a regression model.

[0003] Since the current 2D standard sections are randomly generated and have no correlation with the scanned 3D fetal heart data, if the randomly generated 2D standard sections are used as reference sections, the matching degree between the 2D standard sections generated by the regression model and the 3D fetal heart data is not high. Summary of the Invention

[0004] The present invention aims to solve the technical problems existing in the prior art and provides a 3D fetal heart standard section recognition method, system, electronic device and storage medium, which can overcome the problem of inaccurate generation of 2D standard sections in the prior art.

[0005] According to a first aspect of the present invention, a method for identifying a 3D fetal heart standard section is provided, comprising: Generate a Gaussian noise image based on Gaussian distribution and obtain a 3D fetal heart image; Inputting the Gaussian noise image and the 3D fetal heart image into an image conditional latent U-Net network model, and performing feature fusion on the Gaussian noise image and the 3D fetal heart image; Inputting the fused features into a variational autoencoder (VAE) to generate a 2D standard section of the 3D fetal heart image; The 2D standard section is input into a regression model to obtain the 2D standard section of the 3D fetal heart image.

[0006] On the basis of the above technical solution, the present invention can also make the following improvements.

[0007] Optionally, the image-conditional latent U-Net network model includes a first encoder and a first decoder, inputting the Gaussian noise image and the 3D fetal heart image into the image-conditional latent U-Net network model, and performing feature fusion on the Gaussian noise image and the 3D fetal heart image, including: Downsampling the Gaussian noise image and the 3D fetal heart image based on the first encoder to extract features at different levels; Based on the first decoder, up-sample and fuse the features of different levels of the Gaussian noise image and the 3D fetal heart image extracted by the first encoder to generate fused features.

[0008] Optionally, downsampling the Gaussian noise image and the 3D fetal heart image based on the first encoder to extract features at different levels includes: Downsampling the Gaussian noise image and the 3D fetal heart image respectively by the first encoder, and extracting local features and global features of the Gaussian noise image and the 3D fetal heart image respectively; In which, the first encoder downsamples the Gaussian noise image and the 3D fetal heart image respectively according to preset image condition information, so that the local features and global features extracted from the Gaussian noise image and the 3D fetal heart image meet the condition requirements, and the image condition information includes text prompts, and the text prompts guide the first encoder to generate a specific style of features and / or a target object of interest.

[0009] Optionally, upsampling and fusing features of different levels of the Gaussian noise image and the 3D fetal heart image extracted by the first encoder based on the first decoder to generate fused features includes: According to the local features and global features of the Gaussian noise image and the 3D fetal heart image fused with the image condition information transmitted by the first encoder, the local features and the global features are upsampled and fused to generate a fused feature map that meets the condition requirements.

[0010] Optionally, inputting the Gaussian noise image and the 3D fetal heart image into an image conditional latent U-Net network model, and performing feature fusion on the Gaussian noise image and the 3D fetal heart image includes: Inputting the generated fused feature map and the Gaussian noise image into the image conditional latent U-Net network model again to update the generated fused feature map; Repeat the update multiple times to generate the final fusion feature map.

[0011] Optionally, the variational autoencoder VAE includes a second encoder and a second decoder, inputting the fused features into the variational autoencoder VAE to generate a 2D standard-like section of the 3D fetal heart image, including: The second encoder maps the input fused feature map into a probability distribution in a latent space, where the probability distribution is a Gaussian distribution, and samples a latent variable z from the probability distribution, where the latent variable z is a low-dimensional representation of the fused feature map and contains key information of the fused feature map; The second decoder decodes the latent variable z back to the original data space to generate a reconstructed 2D standard-like slice.

[0012] Optionally, the training process of the variational autoencoder VAE is: Acquire a training sample set, the training sample set including a plurality of samples, each of the samples including a fused feature map and a corresponding 2D class standard section; Training the variational autoencoder (VAE) based on the training sample set to obtain the variational autoencoder (VAE); During training of the variational autoencoder (VAE), the loss of the variational autoencoder (VAE) includes a KL divergence loss of the second encoder and a reconstruction loss of the second decoder, wherein the KL divergence loss represents a difference between a probability distribution generated by the second encoder and a standard normal distribution, and the reconstruction loss represents a difference between a fused feature map output by the second decoder and an actual fused feature map. The parameters of the variational autoencoder (VAE) are adjusted based on the loss to obtain the trained variational autoencoder (VAE).

[0013] According to a second aspect of the present invention, a 3D fetal heart standard section recognition system is provided, comprising: A first generating module is used to generate a Gaussian noise image based on Gaussian distribution and obtain a 3D fetal heart image; a feature fusion module, configured to input the Gaussian noise image and the 3D fetal heart image into an image conditional latent U-Net network model, and perform feature fusion on the Gaussian noise image and the 3D fetal heart image; The second generation module is used to input the fusion features into the variational autoencoder (VAE) to generate a 2D standard section of the 3D fetal heart image; The regression module is used to input the 2D standard section into a regression model and regress to obtain the 2D standard section of the 3D fetal heart image.

[0014] According to a third aspect of the present invention, an electronic device is provided, comprising a memory and a processor, wherein the processor is configured to implement the steps of a 3D fetal heart standard section recognition method when executing a computer management program stored in the memory.

[0015] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, on which a computer management program is stored. When the computer management program is executed by a processor, the steps of the 3D fetal heart standard section recognition method are implemented.

[0016] The present invention provides a 3D fetal heart standard section recognition method, system, electronic device and storage medium, which generates a Gaussian noise image based on Gaussian distribution, adds noise to the captured 3D fetal heart image and generates a 2D-type standard section. The generated 2D-type standard section is generated based on the 3D fetal heart image and has a certain correlation with the 3D fetal heart image. The 2D-type standard section generated in this way is used as a reference section, and the 2D fetal heart section finally obtained is more standard. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flow chart of a 3D fetal heart standard section recognition method provided by the present invention; Figure 2 This is a diagram illustrating the entire framework for 2D standard slice recognition of 3D fetal heart data according to an embodiment of the present invention; Figure 3 Schematic diagram of the working principle of the first encoder according to an embodiment of the present invention; Figure 4 A structural diagram of a 3D fetal heart standard section recognition system provided by an embodiment of the present invention; Figure 5 A schematic diagram of the hardware structure of a possible electronic device provided by the present invention; Figure 6 A schematic diagram of the hardware structure of a possible computer-readable storage medium provided by the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, the technical features in the various embodiments or single embodiments provided by the present invention can be arbitrarily combined with each other to form a feasible technical solution. This combination is not restricted by the sequence of steps and / or structural composition mode, but must be based on the ability of ordinary technicians in this field to implement it. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0019] To address the shortcomings of the existing technology, the present invention proposes a new 3D fetal heart standard section recognition method, which is mainly based on the stable diffusion framework. It first uses 3D fetal heart data to generate an initial 2D-like standard section, and then regresses the 2D-like standard section to obtain the true 2D standard section. The purpose is to improve the accuracy and authenticity of the generated section, and may also include improving the efficiency of the generation process to better meet the demand for 2D standard sections in related fields.

[0020] Figure 1 A flow chart of a 3D fetal heart standard section recognition method provided by the present invention, such as Figure 1 As shown, the method includes: Step 1: Generate a Gaussian noise image based on Gaussian distribution and obtain a 3D fetal heart image.

[0021] See also Figure 2 , which shows the entire framework for 2D standard slice recognition of 3D fetal heart data. In an embodiment of the present invention, noise is added to the acquired 3D fetal heart image. Therefore, a noise image must first be generated. In one embodiment of the present invention, a Gaussian noise image is generated based on a Gaussian distribution. The size of the generated Gaussian noise image is 64*64.

[0022] The scanned 3D fetal heart image is obtained and cut into an image of 77*768 size.

[0023] Step 2: Input the Gaussian noise image and the 3D fetal heart image into an image conditional latent U-Net network model, and perform feature fusion on the Gaussian noise image and the 3D fetal heart image.

[0024] It can be understood that the generated Gaussian noise image and the 3D fetal heart image are jointly input into the image conditional latent U-Net network model for feature fusion.

[0025] In a possible implementation of the present invention, the image-conditional latent U-Net network model includes a first encoder and a first decoder, inputs the Gaussian noise image and the 3D fetal heart image into the image-conditional latent U-Net network model, and performs feature fusion on the Gaussian noise image and the 3D fetal heart image, including: Step 21 : Downsample the Gaussian noise image and the 3D fetal heart image based on the first encoder to extract features at different levels.

[0026] The image-conditioned latent U-Net model primarily consists of a first encoder and a first decoder. The first encoder primarily extracts features from the Gaussian noise image and the 3D fetal heart image. During feature extraction, the first encoder primarily downsamples the Gaussian noise image and the 3D fetal heart image to extract features at different levels of the image and the 3D fetal heart image, respectively.

[0027] In a possible implementation of the present invention, downsampling the Gaussian noise image and the 3D fetal heart image based on the first encoder to extract features at different levels includes: The Gaussian noise image and the 3D fetal heart image are downsampled by the first encoder, and local features and global features of the Gaussian noise image and the 3D fetal heart image are extracted respectively.

[0028] In which, the first encoder downsamples the Gaussian noise image and the 3D fetal heart image respectively according to preset image condition information, so that the local features and global features extracted from the Gaussian noise image and the 3D fetal heart image meet the condition requirements, and the image condition information includes text prompts, and the text prompts guide the encoder to generate a specific style of features and / or a target object of attention.

[0029] It is understandable that when the first encoder is used to extract the features of the Gaussian noise image and the 3D fetal heart image respectively, preset image condition information is provided to the first encoder, for example, prompting the first encoder to focus on which area and which target object's features to extract, and specifying the style of the feature map extracted by the first encoder, etc. In this way, the feature map extracted by the first encoder will meet certain conditions for subsequent fusion.

[0030] Step 22: upsample and fuse the features of the Gaussian noise image and the 3D fetal heart image at different levels extracted by the encoder based on the first decoder to generate fused features.

[0031] In a possible implementation of the present invention, upsampling and fusing features of different levels of the Gaussian noise image and the 3D fetal heart image extracted by the first encoder based on the first decoder to generate fused features includes: According to the local features and global features of the Gaussian noise image and the 3D fetal heart image fused with the image condition information transmitted by the first encoder, the local features and the global features are upsampled and fused to generate a fused feature map that meets the condition requirements.

[0032] It is understandable that the first encoder in step 21 extracts the local features and global features of the Gaussian noise image and the 3D fetal heart image of the image condition information, and the first decoder upsamples and fuses the local features and global features of the Gaussian noise image and the 3D fetal heart image to generate fused features.

[0033] In one possible implementation of the present invention, inputting the Gaussian noise image and the 3D fetal heart image into an image conditional latent U-Net network model, and performing feature fusion on the Gaussian noise image and the 3D fetal heart image includes: The generated fused feature map and the Gaussian noise image are input into the image conditional latent U-Net again to update the generated fused feature map; the fusion update is repeated multiple times to generate the final fused feature map.

[0034] It is understandable that after the features of the Gaussian noise image and the 3D fetal heart image are initially fused to generate fused features, the initially generated fused features and the features of the Gaussian noise image are fused again, and the fusion is performed multiple times to generate the final fused feature map.

[0035] Step 3: Input the fused features into a variational autoencoder (VAE) to generate a 2D standard section of the 3D fetal heart image.

[0036] It is understandable that after step 2 generates the fusion feature map, this step inputs the fusion feature map into the variational autoencoder VAE to generate a 2D standard section of the 3D fetal heart image. The variational autoencoder VAE includes a second encoder and a second decoder. Figure 3 , the second encoder maps the input fusion feature map into a probability distribution of the latent space, which is a Gaussian distribution, and samples a latent variable z from the probability distribution, wherein the latent variable z is a low-dimensional representation of the fusion feature map and contains the key information of the fusion feature map; the second decoder decodes the latent variable z back to the original data space x'=d(z) to generate a reconstructed 2D class standard section.

[0037] The variational autoencoder (VAE) is trained through self-supervised learning. The training process of the variational autoencoder (VAE) is as follows: Acquire a training sample set, the training sample set including a plurality of samples, each of the samples including a fused feature map and a corresponding 2D class standard section; Training the variational autoencoder (VAE) based on the training sample set to obtain the variational autoencoder (VAE); During training of the variational autoencoder (VAE), the loss of the variational autoencoder (VAE) includes a KL divergence loss of the second encoder and a reconstruction loss of the second decoder, wherein the KL divergence loss represents a difference between a probability distribution generated by the second encoder and a standard normal distribution, and the reconstruction loss represents a difference between a fused feature map output by the second decoder and an actual fused feature map. The parameters of the variational autoencoder (VAE) are adjusted based on the loss to obtain the trained variational autoencoder (VAE).

[0038] It is understandable that training a variational autoencoder (VAE) primarily involves training the fusion feature map to a 2D standard cross-section. The fused feature map generated in step 2 is fed into the trained variational autoencoder (VAE) to output a 2D standard cross-section of the 3D fetal heart image.

[0039] Step 4: Input the 2D standard section into a regression model to regress and obtain the 2D standard section of the 3D fetal heart image.

[0040] It can be understood that the regression model learns the conversion parameters from 2D standard-like slices to 2D standard slices through the regression task. During the regression process, the 2D standard-like slices are converted using the initial conversion parameters. The resulting 2D standard slices are compared with the actual 2D standard slices, and the conversion parameters of the regression model are adjusted based on the difference between the two. The 2D standard-like slices of the 3D fetal heart image generated in step 3 are then converted to 2D standard slices based on the regression model.

[0041] See also Figure 4 , is a structural diagram of a 3D fetal heart standard section recognition system provided by an embodiment of the present invention, the recognition system includes: A first generating module 401 is configured to generate a Gaussian noise image based on a Gaussian distribution and acquire a 3D fetal heart image; a feature fusion module 402 for inputting the Gaussian noise image and the 3D fetal heart image into an image conditional latent U-Net network model to perform feature fusion on the Gaussian noise image and the 3D fetal heart image; The second generating module 403 is configured to input the fused features into a variational autoencoder (VAE) to generate a 2D standard section of the 3D fetal heart image; The regression module 404 is configured to input the 2D standard section into a regression model to regress and obtain the 2D standard section of the 3D fetal heart image.

[0042] It can be understood that the 3D fetal heart standard section recognition system provided by the present invention corresponds to the 3D fetal heart standard section recognition method provided in the aforementioned embodiments. The relevant technical features of the 3D fetal heart standard section recognition system can refer to the relevant technical features of the 3D fetal heart standard section recognition method, which will not be repeated here.

[0043] See also Figure 5 , Figure 5 Schematic diagram of an embodiment of an electronic device provided by an embodiment of the present invention. Figure 5 As shown, an embodiment of the present invention provides an electronic device 500, including a memory 510, a processor 520, and a computer program 511 stored in the memory 510 and executable on the processor 520. When the processor 520 executes the computer program 511, the following steps are implemented: generating a Gaussian noise image based on a Gaussian distribution, and acquiring a 3D fetal heart image; inputting the Gaussian noise image and the 3D fetal heart image into an image conditional latent U-Net to generate a 2D standard cross-section of the 3D fetal heart image; and inputting the 2D standard cross-section into a variational autoencoder VAE to generate a 2D standard cross-section of the 3D fetal heart image.

[0044] See also Figure 6 , Figure 6 Schematic diagram of an embodiment of a computer-readable storage medium provided by the present invention. Figure 6 As shown, this embodiment provides a computer-readable storage medium 600, on which a computer program 611 is stored. When the computer program 611 is executed by a processor, the following steps are implemented: generating a Gaussian noise image based on a Gaussian distribution and obtaining a 3D fetal heart image; inputting the Gaussian noise image and the 3D fetal heart image into an image conditional latent U-Net to generate a 2D class standard section of the 3D fetal heart image; inputting the 2D class standard section into a variational autoencoder VAE to generate a 2D standard section of the 3D fetal heart image.

[0045] The embodiments of the present invention provide a 3D fetal heart standard section recognition method, system, electronic device and storage medium, which generate a Gaussian noise image based on Gaussian distribution, perform noise processing on the captured 3D fetal heart image and generate a 2D-type standard section. The generated 2D-type standard section is generated based on the 3D fetal heart image and has a certain correlation with the 3D fetal heart image. The 2D-type standard section generated in this way is used as a reference section, and the final 2D fetal heart section obtained is more standard.

[0046] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0047] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0048] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0049] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0050] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0051] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0052] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A 3D fetal heart standard section recognition method, characterized in that: include: Generate a Gaussian noise image based on Gaussian distribution and obtain a 3D fetal heart image; Inputting the Gaussian noise image and the 3D fetal heart image into an image conditional latent U-Net network model, and performing feature fusion on the Gaussian noise image and the 3D fetal heart image; Inputting the fused features into a variational autoencoder (VAE) to generate a 2D standard section of the 3D fetal heart image; The 2D standard section is input into a regression model to obtain the 2D standard section of the 3D fetal heart image.

2. The 3D fetal heart standard section recognition method according to claim 1, characterized in that: The image-conditional latent U-Net network model includes a first encoder and a first decoder, inputs the Gaussian noise image and the 3D fetal heart image into the image-conditional latent U-Net network model, and performs feature fusion on the Gaussian noise image and the 3D fetal heart image, including: Downsampling the Gaussian noise image and the 3D fetal heart image based on the first encoder to extract features at different levels; Based on the first decoder, up-sample and fuse the features of different levels of the Gaussian noise image and the 3D fetal heart image extracted by the first encoder to generate fused features.

3. The 3D fetal heart standard section recognition method according to claim 2, characterized in that: The downsampling of the Gaussian noise image and the 3D fetal heart image based on the first encoder to extract features at different levels includes: Downsampling the Gaussian noise image and the 3D fetal heart image respectively by the first encoder, and extracting local features and global features of the Gaussian noise image and the 3D fetal heart image respectively; In which, the first encoder downsamples the Gaussian noise image and the 3D fetal heart image respectively according to preset image condition information, so that the local features and global features extracted from the Gaussian noise image and the 3D fetal heart image meet the condition requirements, and the image condition information includes text prompts, and the text prompts guide the first encoder to generate a specific style of features and / or a target object of interest.

4. The 3D fetal heart standard section recognition method according to claim 3, characterized in that: The upsampling and fusing the features of the Gaussian noise image and the 3D fetal heart image at different levels extracted by the first encoder based on the first decoder to generate fused features includes: According to the local features and global features of the Gaussian noise image and the 3D fetal heart image fused with the image condition information transmitted by the first encoder, the local features and the global features are upsampled and fused to generate a fused feature map that meets the condition requirements.

5. The 3D fetal heart standard section recognition method according to claim 1, characterized in that: Inputting the Gaussian noise image and the 3D fetal heart image into an image conditional latent U-Net network model, and performing feature fusion on the Gaussian noise image and the 3D fetal heart image, including: Inputting the generated fused feature map and the Gaussian noise image into the image conditional latent U-Net network model again to update the generated fused feature map; Repeat the update multiple times to generate the final fusion feature map.

6. The 3D fetal heart standard section recognition method according to claim 1, characterized in that: The variational autoencoder VAE includes a second encoder and a second decoder, inputs the fusion features into the variational autoencoder VAE, and generates a 2D standard section of the 3D fetal heart image, including: The second encoder maps the input fused feature map into a probability distribution in a latent space, where the probability distribution is a Gaussian distribution, and samples a latent variable z from the probability distribution, where the latent variable z is a low-dimensional representation of the fused feature map and contains key information of the fused feature map; The second decoder decodes the latent variable z back to the original data space to generate a reconstructed 2D standard-like slice.

7. The 3D fetal heart standard section recognition method according to claim 6, characterized in that: The training process of the variational autoencoder VAE is: Acquire a training sample set, the training sample set including a plurality of samples, each of the samples including a fused feature map and a corresponding 2D class standard section; Training the variational autoencoder (VAE) based on the training sample set to obtain the variational autoencoder (VAE); During training of the variational autoencoder (VAE), the loss of the variational autoencoder (VAE) includes a KL divergence loss of the second encoder and a reconstruction loss of the second decoder, wherein the KL divergence loss represents a difference between a probability distribution generated by the second encoder and a standard normal distribution, and the reconstruction loss represents a difference between a fused feature map output by the second decoder and an actual fused feature map. The parameters of the variational autoencoder (VAE) are adjusted based on the loss to obtain the trained variational autoencoder (VAE).

8. A 3D fetal heart standard section recognition system, characterized in that: include: A first generating module is used to generate a Gaussian noise image based on Gaussian distribution and obtain a 3D fetal heart image; a feature fusion module, configured to input the Gaussian noise image and the 3D fetal heart image into an image conditional latent U-Net network model, and perform feature fusion on the Gaussian noise image and the 3D fetal heart image; The second generation module is used to input the fusion features into the variational autoencoder (VAE) to generate a 2D standard section of the 3D fetal heart image; The regression module is used to input the 2D standard section into a regression model and regress to obtain the 2D standard section of the 3D fetal heart image.

9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the processor is used to implement the steps of the 3D fetal heart standard section recognition method according to any one of claims 1 to 7 when executing a computer management program stored in the memory.

10. A computer-readable storage medium, characterized in that A computer management program is stored thereon, and when the computer management program is executed by the processor, the steps of the 3D fetal heart standard section recognition method according to any one of claims 1 to 7 are implemented.

Images