A three-dimensional modeling method and system for medical imaging data

Abstract

The present invention belongs to the field of medical image processing technology and discloses a method and system for three-dimensional modeling of medical image data. The method comprises the following steps: constructing an image segmentation model and a three-dimensional reconstruction model based on a number of historical medical image data and corresponding historical patient diagnostic data using a deep learning algorithm; acquiring real-time medical image data and corresponding real-time patient diagnostic data, and using the image segmentation model to segment the real-time medical image data to obtain a real-time segmented image; and using a three-dimensional reconstruction model to perform three-dimensional reconstruction of the real-time segmented image and real-time patient diagnostic data to obtain a real-time three-dimensional model of the real-time medical image data. The present invention solves the problems of insufficient segmentation accuracy, distorted three-dimensional reconstruction, and low computational efficiency in the prior art.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical image processing, and in particular relates to a three-dimensional modeling method and system for medical image data. Background Art

[0002] Medical imaging data refers to image data acquired through imaging technology during medical diagnosis and treatment. This data is typically generated by various medical imaging devices, such as X-rays, computed tomography (CT), magnetic resonance imaging (MRI), ultrasound, and positron emission tomography (PET). Three-dimensional modeling of medical imaging data is a crucial component of medical image analysis and is crucial for disease diagnosis, treatment, and prognosis.

[0003] Existing 3D modeling of medical imaging data has the following defects:

[0004] 1) Insufficient segmentation accuracy: Existing image segmentation algorithms cannot accurately segment complex medical image data, especially when processing images with high noise and low contrast.

[0005] 2) 3D reconstruction distortion: Existing 3D reconstruction technologies introduce distortion during the reconstruction process. Especially when dealing with non-uniform or irregular structures, the reconstructed 3D model may differ significantly from the actual anatomical structure.

[0006] 3) Low computational efficiency: Existing 3D modeling methods require a large amount of computing resources, resulting in slow processing speed, which is not conducive to rapid clinical diagnosis and treatment planning. Summary of the Invention

[0007] In order to solve the problems of insufficient segmentation accuracy, 3D reconstruction distortion and low computational efficiency in the prior art, the present invention aims to provide a 3D modeling method and system for medical imaging data.

[0008] The technical solution adopted in the present invention is:

[0009] A three-dimensional modeling method for medical imaging data comprises the following steps:

[0010] Based on a number of historical medical imaging data and corresponding historical patient diagnosis data, a deep learning algorithm is used to build an image segmentation model and a 3D reconstruction model;

[0011] Collect real-time medical image data and corresponding real-time patient diagnostic data, and use an image segmentation model to perform image segmentation on the real-time medical image data to obtain a real-time segmented image;

[0012] The real-time segmented image and the real-time patient diagnostic data are 3D reconstructed using the 3D reconstruction model to obtain a real-time 3D model of the real-time medical imaging data.

[0013] Furthermore, based on a number of historical medical imaging data and corresponding historical patient diagnosis data, a deep learning algorithm is used to construct an image segmentation model and a 3D reconstruction model, including the following steps:

[0014] Collecting a number of historical medical image data and corresponding historical patient diagnosis data, and preprocessing them to obtain a number of preprocessed historical medical image data and a number of preprocessed historical patient diagnosis data;

[0015] Based on a number of pre-processed historical medical imaging data, a deep learning algorithm is used to build an image segmentation model and generate a number of historical segmentation images;

[0016] Based on several historical segmentation images and corresponding pre-processed historical patient diagnosis data, a deep learning algorithm is used to construct a three-dimensional reconstruction model.

[0017] Furthermore, the preprocessing of historical medical imaging data includes data cleaning, Gaussian denoising, size normalization, image enhancement, and labeling in sequence;

[0018] The preprocessing of historical patient diagnostic data includes data cleaning, Gaussian denoising, and magnitude normalization in sequence.

[0019] Furthermore, the image segmentation model is constructed based on the CNN-FPN-SAM-Med2D algorithm, and the image segmentation model includes a graph feature extraction module based on the CNN algorithm, a feature fusion module based on the FPN algorithm, and an image segmentation module constructed based on the SAM-Med2D algorithm, which are connected in sequence.

[0020] Furthermore, based on a number of pre-processed historical medical image data, a deep learning algorithm is used to construct an image segmentation model and generate a number of historical segmentation images, including the following steps:

[0021] Using the CNN algorithm, we build the basic network structure of the image segmentation model and obtain the initial graph feature extraction module. The basic network structure consists of several convolutional layers and several pooling layers that are connected alternately in sequence.

[0022] Add skip connections to extract feature maps of all convolutional layers in the basic network structure, and use upsampling and lateral connections based on several feature maps to build a feature pyramid and obtain the initial feature fusion module;

[0023] The initial image segmentation module is constructed using the SAM-Med2D algorithm, and the input end of the initial image segmentation module is connected to the output end of the initial feature fusion module;

[0024] Integrate the initial graph feature extraction module, the initial feature fusion module and the initial image segmentation module to obtain the initial image segmentation model;

[0025] The pre-processed historical medical imaging data were divided into a model training set and a model test set in a ratio of 7:3;

[0026] Input the model training set, optimize the initial image segmentation model and train it to obtain the optimized image segmentation model, and generate several historical segmentation images;

[0027] Input the model test set, perform model testing on the optimized image segmentation model, and obtain the model test accuracy. If the model test accuracy is greater than the accuracy threshold, the final image segmentation model is output.

[0028] Furthermore, the three-dimensional reconstruction model is constructed based on the cGAN-MLP algorithm, and the three-dimensional reconstruction model includes a conditional information embedder and a conditional information processor both constructed based on the MLP algorithm, and a generator and a discriminator both constructed based on the RNN algorithm. The generator is connected to the conditional information embedder and the discriminator respectively, and the discriminator is connected to the conditional information processor.

[0029] Furthermore, based on several historical segmentation images and corresponding pre-processed historical patient diagnosis data, a deep learning algorithm is used to construct a 3D reconstruction model, including the following steps:

[0030] Setting corresponding real three-dimensional data for each historical segmentation image and corresponding pre-processed historical patient diagnosis data;

[0031] Use the cGAN algorithm to build the initial generator and initial discriminator, and use the MLP algorithm to set the conditional information embedder for the initial generator and the conditional information processor for the initial discriminator;

[0032] Integrate the initial generator, initial discriminator, conditional information embedder, and conditional information processor to obtain the initial 3D reconstruction model;

[0033] Combine the first loss function of the initial generator and the second loss function of the initial discriminator to obtain a comprehensive loss function;

[0034] Extracting historical image features of historical segmented images and historical data features of pre-processed historical patient diagnosis data;

[0035] Use the conditional information embedder to conditionally embed historical image features and historical data features to obtain historical conditional information embedding features;

[0036] Using the initial generator, embedding features based on historical condition information to generate three-dimensional data, thereby generating three-dimensional data;

[0037] Using a condition information processor, condition information processing is performed based on the real three-dimensional data and the corresponding generated three-dimensional data to obtain historical condition information;

[0038] Using the initial discriminator, data discrimination is performed based on the real 3D data, the corresponding generated 3D data, and the historical condition information to obtain the historical data discrimination result;

[0039] Traverse all historical segmentation images and the corresponding pre-processed historical patient diagnosis data, repeat the above steps, and optimize the initial 3D reconstruction model;

[0040] Use the comprehensive loss function to generate the historical loss value of the optimization training process. If the historical loss value is lower than the loss value threshold, the final 3D reconstruction model is output; otherwise, the optimization training continues.

[0041] Furthermore, real-time medical image data and corresponding real-time patient diagnostic data are collected, and an image segmentation model is used to perform image segmentation on the real-time medical image data to obtain a real-time segmented image, including the following steps:

[0042] Collecting real-time medical imaging data and corresponding real-time patient diagnostic data, and inputting the real-time medical imaging data into an image segmentation model;

[0043] Use the graph feature extraction module of the image segmentation model to extract several real-time feature graphs of real-time medical imaging data;

[0044] Use the feature fusion module of the image segmentation model to fuse several real-time feature maps to obtain real-time fusion features;

[0045] The image segmentation module of the image segmentation model is used to perform image segmentation on the real-time fusion features to obtain a real-time segmented image.

[0046] Furthermore, using the 3D reconstruction model, the real-time segmented image and the real-time patient diagnostic data are 3D reconstructed to obtain a real-time 3D model of the real-time medical imaging data, including the following steps:

[0047] inputting the real-time segmentation image and the real-time patient diagnosis data into the three-dimensional reconstruction model, and extracting the real-time image features of the real-time segmentation image and the real-time data features of the real-time patient diagnosis data;

[0048] Using the conditional information embedder of the 3D reconstruction model, conditionally embed the real-time image features and real-time data features to obtain real-time conditional information embedding features;

[0049] Using a 3D reconstruction model generator, embed features based on real-time condition information to generate 3D data and obtain real-time 3D data;

[0050] Based on the real-time three-dimensional data, three-dimensional reconstruction is performed to obtain a real-time three-dimensional model of the real-time medical imaging data.

[0051] A three-dimensional modeling system for medical image data is used to implement a three-dimensional modeling method. The system includes a model building unit, an image segmentation unit and a three-dimensional reconstruction unit connected in sequence.

[0052] The beneficial effects of the present invention are:

[0053] The present invention discloses a 3D modeling method and system for medical image data. The constructed image segmentation model can more accurately segment complex medical image data, reduce modeling errors caused by inaccurate segmentation, and improve segmentation accuracy. The constructed 3D reconstruction model combines real-time segmented images and real-time patient diagnostic data for 3D reconstruction, which can more realistically restore anatomical structures and reduce distortion problems in the reconstruction process. By segmenting two-dimensional images and restoring three-dimensional structures from the segmented images, the processing speed and computing efficiency are improved, so that 3D modeling can be completed quickly, which is particularly suitable for clinical scenarios that require rapid response. Automated image segmentation and 3D reconstruction are performed with a high degree of automation, which reduces manual intervention and reduces error rates.

[0054] Other beneficial effects of the present invention will be further described in the specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a flowchart of the three-dimensional modeling method of medical image data in the present invention.

[0056] Figure 2 It is a structural block diagram of the three-dimensional modeling system of medical imaging data in the present invention. DETAILED DESCRIPTION

[0057] The present invention will be further explained below with reference to the accompanying drawings and specific embodiments.

[0058] Example 1:

[0059] like Figure 1 As shown, this embodiment provides a three-dimensional modeling method for medical image data, including the following steps:

[0060] S1: Based on a number of historical medical imaging data and corresponding historical patient diagnosis data, a deep learning algorithm is used to construct an image segmentation model and a 3D reconstruction model, including the following steps:

[0061] S1-1: Collecting a number of historical medical imaging data and corresponding historical patient diagnosis data, and preprocessing them to obtain a number of preprocessed historical medical imaging data and a number of preprocessed historical patient diagnosis data;

[0062] The preprocessing of historical medical imaging data includes data cleaning, Gaussian denoising, size normalization, image enhancement, and labeling in sequence;

[0063] Data cleaning is used to remove incomplete and low-quality image data; identify and correct errors in the data, such as incorrect labels or metadata; and remove duplicate image data;

[0064] Gaussian denoising: Applying a Gaussian filter to smooth image data to reduce random noise and improve image quality; this helps the model better learn the true characteristics of the image;

[0065] Size normalization adjusts all image data to a uniform size for easier model processing; this helps reduce the model's dependence on image size;

[0066] Image enhancement, which applies a series of image processing techniques (such as rotation, scaling, flipping, contrast adjustment, etc.) to expand the dataset, improve the generalization ability of the model, simulate various imaging conditions that may be encountered in reality, and make the model more robust;

[0067] Labeling: labeling image data to mark areas of interest (such as diseased tissue); this usually requires the participation of professionals to ensure the accuracy and consistency of labeling;

[0068] The preprocessing of historical patient diagnostic data includes data cleaning, Gaussian denoising, and magnitude normalization in sequence;

[0069] Data cleaning: removing invalid, erroneous, or missing data; ensuring data integrity and accuracy; Gaussian denoising: For numerical data, which may contain measurement errors or outliers, Gaussian denoising can reduce the impact of these noises; in diagnostic data, Gaussian denoising may involve processing statistical outliers; magnitude normalization: scaling feature data of different magnitudes to a small, uniform interval, such as [0, 1] or [-1, 1]; this helps prevent certain features from dominating model training due to their large numerical range;

[0070] S1-2: Based on some pre-processed historical medical imaging data, use deep learning algorithms to build an image segmentation model and generate some historical segmentation images;

[0071] The image segmentation model is based on the Convolutional Neural Networks (CNN)-Feature Pyramid Networks (FPN)-Segment Anything Model-Medicine 2Dimensionality (SAM-Med2D) algorithm. The model includes a graph feature extraction module based on the CNN algorithm, a feature fusion module based on the FPN algorithm, and an image segmentation module based on the SAM-Med2D algorithm.

[0072] The graph feature extraction module extracts hierarchical features from images through a series of convolutional layers, activation functions, and pooling layers. The feature fusion module uses skip connections to fuse feature maps from different layers extracted by CNN. Through upsampling and lateral connections, it constructs a feature pyramid, transferring high-level semantic information to lower layers and enhancing the semantic expression ability of low-level features. This effectively combines low-level detail features with high-level semantic features, allowing the network to maintain a high accuracy rate when detecting tissues or lesions of different sizes in medical imaging data. SAM-Med2D uses an attention mechanism and a conditional generation model to predict segmentation masks in images, which are used to generate segmented images.

[0073] Based on some pre-processed historical medical imaging data, a deep learning algorithm is used to build an image segmentation model and generate several historical segmentation images, including the following steps:

[0074] S1-2-1: Use the CNN algorithm to build the basic network structure of the image segmentation model and obtain the initial graph feature extraction module; the basic network structure includes several convolutional layers and several pooling layers connected alternately in sequence;

[0075] The convolution layer is used to extract local features, and the pooling layer is used to reduce the spatial dimension of the features. This structure can effectively extract key features in the image and provide rich information for subsequent segmentation tasks;

[0076] S1-2-2: Add skip connections to extract feature maps of all convolutional layers in the basic network structure, and use upsampling and lateral connections based on several feature maps to build a feature pyramid and obtain the initial feature fusion module;

[0077] Skip connections (such as residual connections) allow direct information propagation within the network, reducing the vanishing gradient problem. Feature pyramids fuse feature maps of different scales through upsampling and lateral connections to preserve detailed information. Feature pyramid networks can improve the network's utilization of multi-scale features, helping segmentation models more accurately identify objects of different sizes.

[0078] S1-2-3: Use the SAM-Med2D algorithm to construct an initial image segmentation module, and connect the input end of the initial image segmentation module to the output end of the initial feature fusion module;

[0079] The SAM-Med2D algorithm can improve the accuracy and robustness of segmentation, especially when processing complex medical images;

[0080] S1-2-4: Integrate the initial graph feature extraction module, the initial feature fusion module, and the initial image segmentation module to obtain the initial image segmentation model;

[0081] S1-2-5: Divide the pre-processed historical medical imaging data into a model training set and a model test set in a ratio of 7:3;

[0082] S1-2-6: Input the model training set, optimize the initial image segmentation model, obtain the optimized image segmentation model, and generate several historical segmentation images;

[0083] S1-2-7: Input the model test set, perform model testing on the optimized image segmentation model, and obtain the model test accuracy. If the model test accuracy is greater than the accuracy threshold, the final image segmentation model is output;

[0084] S1-3: Based on several historical segmentation images and corresponding pre-processed historical patient diagnosis data, a deep learning algorithm is used to construct a 3D reconstruction model;

[0085] The 3D reconstruction model is built based on the Conditional Generative Adversarial Network (cGAN)-Multilayer Perceptron (MLP) algorithm. The 3D reconstruction model includes a conditional information embedder and conditional information processor, both built based on the MLP algorithm, as well as a generator and discriminator, both built based on the RNN algorithm. The generator is connected to the conditional information embedder and discriminator, respectively, and the discriminator is connected to the conditional information processor.

[0086] The conditional information embedder is used to process conditional information including image features and data features to obtain conditional information embedding features in a sequence format, thereby integrating multimodal conditional information into the generation process. The generator is used to generate three-dimensional data based on the conditional information embedding features and random noise output by the conditional information embedder. The conditional information processor is used to process additional conditional information of the three-dimensional data to help the discriminator more accurately judge the authenticity of the three-dimensional data. The discriminator analyzes whether the generated three-dimensional data is authentic and meets the given conditional information. The generator and the discriminator compete with each other through an adversarial training process. The generator attempts to generate three-dimensional data that can deceive the discriminator, while the discriminator attempts to better distinguish between real and fake three-dimensional data, thereby achieving accurate and efficient generation of three-dimensional data.

[0087] Based on several historical segmentation images and the corresponding pre-processed historical patient diagnosis data, a deep learning algorithm is used to construct a 3D reconstruction model, including the following steps:

[0088] S1-3-1: Set corresponding real 3D data for each historical segmentation image and the corresponding pre-processed historical patient diagnosis data;

[0089] Set real 3D data as reference standard for each historical segmentation image and the corresponding pre-processed patient diagnosis data;

[0090] S1-3-2: Use the cGAN algorithm to build the initial generator and initial discriminator, and use the MLP algorithm to set the conditional information embedder for the initial generator and the conditional information processor for the initial discriminator;

[0091] S1-3-3: Integrate the initial generator, initial discriminator, conditional information embedder, and conditional information processor to obtain the initial 3D reconstruction model;

[0092] S1-3-4: Combine the first loss function of the initial generator and the second loss function of the initial discriminator to obtain a comprehensive loss function;

[0093] The comprehensive loss function can be used to optimize the model more comprehensively and balance the training of the generator and discriminator;

[0094] S1-3-5: Extract historical image features of historical segmented images and historical data features of pre-processed historical patient diagnosis data;

[0095] S1-3-6: Use the conditional information embedder to conditionally embed historical image features and historical data features to obtain historical conditional information embedding features;

[0096] S1-3-7: Using the initial generator, embedding features based on historical condition information to generate three-dimensional data, thereby obtaining generated three-dimensional data;

[0097] S1-3-8: Using a condition information processor, perform condition information processing based on the real 3D data and the corresponding generated 3D data to obtain historical condition information;

[0098] S1-3-9: Using the initial discriminator, perform data discrimination based on the real 3D data, the corresponding generated 3D data, and the historical condition information to obtain the historical data discrimination result;

[0099] S1-3-10: Traverse all historical segmentation images and the corresponding pre-processed historical patient diagnosis data, repeat the above steps, and optimize the initial 3D reconstruction model;

[0100] S1-3-11: Use the comprehensive loss function to generate the historical loss value of the optimization training process. If the historical loss value is lower than the loss value threshold, the final 3D reconstruction model is output; otherwise, the optimization training continues;

[0101] 3D reconstruction models can generate realistic 3D data using patient diagnostic data and segmented images. They have broad application potential in medical imaging, surgical planning, disease diagnosis, and other fields, and can improve the quality and efficiency of medical services.

[0102] S2: Collecting real-time medical image data and corresponding real-time patient diagnostic data, and using an image segmentation model to perform image segmentation on the real-time medical image data to obtain a real-time segmented image, including the following steps:

[0103] S2-1: Collect real-time medical imaging data and corresponding real-time patient diagnostic data, and input the real-time medical imaging data into the image segmentation model;

[0104] S2-2: Use the graph feature extraction module of the image segmentation model to extract several real-time feature graphs of real-time medical imaging data;

[0105] S2-3: Use the feature fusion module of the image segmentation model to fuse several real-time feature maps to obtain real-time fusion features;

[0106] S2-4: Use the image segmentation module of the image segmentation model to perform image segmentation on the real-time fusion features to obtain a real-time segmented image;

[0107] The image segmentation module uses the fused features to perform segmentation tasks and outputs segmentation masks that mark the areas of interest in the image;

[0108] S3: Using the 3D reconstruction model, 3D reconstruction is performed on the real-time segmented image and the real-time patient diagnostic data to obtain a real-time 3D model of the real-time medical imaging data, including the following steps:

[0109] S3-1: inputting the real-time segmentation image and the real-time patient diagnosis data into the 3D reconstruction model, and extracting the real-time image features of the real-time segmentation image and the real-time data features of the real-time patient diagnosis data;

[0110] Feature extraction is the basis for understanding the input data and provides the necessary information for 3D reconstruction, ensuring that the reconstruction process can take into account the details of the image and the patient's specific diagnostic information;

[0111] S3-2: Use the conditional information embedder of the 3D reconstruction model to conditionally embed the real-time image features and real-time data features to obtain real-time conditional information embedding features;

[0112] By embedding conditional information, the generator can be guided to more accurately generate 3D data corresponding to specific diagnoses and image features, improving the reconstructed objectivity and accuracy.

[0113] S3-3: Using the generator of the 3D reconstruction model, embedding features based on real-time condition information to generate 3D data and obtain real-time 3D data;

[0114] Generating 3D data is the core of the reconstruction process. Recovering 3D structures from 2D images is crucial for understanding complex anatomical structures and pathological conditions.

[0115] S3-4: Perform three-dimensional reconstruction based on the real-time three-dimensional data to obtain a real-time three-dimensional model of the real-time medical imaging data.

[0116] Example 2:

[0117] like Figure 2 As shown, this embodiment provides a three-dimensional modeling system for medical image data, which is used to implement a three-dimensional modeling method. The system includes a model building unit, an image segmentation unit, and a three-dimensional reconstruction unit connected in sequence;

[0118] A model building unit, configured to build an image segmentation model and a three-dimensional reconstruction model using a deep learning algorithm based on a number of historical medical imaging data and corresponding historical patient diagnosis data;

[0119] An image segmentation unit is used to collect real-time medical image data and corresponding real-time patient diagnostic data, and use an image segmentation model to perform image segmentation on the real-time medical image data to obtain a real-time segmented image;

[0120] The three-dimensional reconstruction unit is used to use the three-dimensional reconstruction model to perform three-dimensional reconstruction on the real-time segmented image and the real-time patient diagnostic data to obtain a real-time three-dimensional model of the real-time medical imaging data.

[0121] The present invention discloses a 3D modeling method and system for medical image data. The constructed image segmentation model can more accurately segment complex medical image data, reduce modeling errors caused by inaccurate segmentation, and improve segmentation accuracy. The constructed 3D reconstruction model combines real-time segmented images and real-time patient diagnostic data for 3D reconstruction, which can more realistically restore anatomical structures and reduce distortion problems in the reconstruction process. By segmenting two-dimensional images and restoring three-dimensional structures from the segmented images, the processing speed and computing efficiency are improved, so that 3D modeling can be completed quickly, which is particularly suitable for clinical scenarios that require rapid response. Automated image segmentation and 3D reconstruction are performed with a high degree of automation, which reduces manual intervention and reduces error rates.

[0122] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A three-dimensional modeling method for medical image data, characterized by: The steps include: Based on a number of historical medical imaging data and corresponding historical patient diagnosis data, a deep learning algorithm is used to build an image segmentation model and a 3D reconstruction model; The image segmentation model is constructed based on the CNN-FPN-SAM-Med2D algorithm, and the image segmentation model includes a graph feature extraction module based on the CNN algorithm, a feature fusion module based on the FPN algorithm, and an image segmentation module constructed by the SAM-Med2D algorithm, which are connected in sequence; Based on some pre-processed historical medical imaging data, a deep learning algorithm is used to build an image segmentation model and generate several historical segmentation images, including the following steps: Using the CNN algorithm, a basic network structure of the image segmentation model is constructed to obtain an initial graph feature extraction module; the basic network structure includes several convolutional layers and several pooling layers that are alternately connected in sequence; Add skip connections to extract feature maps of all convolutional layers in the basic network structure, and use upsampling and lateral connections based on several feature maps to build a feature pyramid and obtain the initial feature fusion module; An initial image segmentation module is constructed using the SAM-Med2D algorithm, and the input end of the initial image segmentation module is connected to the output end of the initial feature fusion module; Integrate the initial graph feature extraction module, the initial feature fusion module and the initial image segmentation module to obtain the initial image segmentation model; The pre-processed historical medical imaging data were divided into a model training set and a model test set in a ratio of 7:3; Input the model training set, optimize the initial image segmentation model and train it to obtain the optimized image segmentation model, and generate several historical segmentation images; Input the model test set, perform model testing on the optimized image segmentation model, and obtain the model test accuracy. If the model test accuracy is greater than the accuracy threshold, the final image segmentation model is output; Collect real-time medical image data and corresponding real-time patient diagnostic data, and use an image segmentation model to perform image segmentation on the real-time medical image data to obtain a real-time segmented image; The real-time segmented image and the real-time patient diagnostic data are 3D reconstructed using the 3D reconstruction model to obtain a real-time 3D model of the real-time medical imaging data.

2. The three-dimensional modeling method of medical image data according to claim 1, characterized in that: Based on a number of historical medical imaging data and corresponding historical patient diagnosis data, a deep learning algorithm is used to build an image segmentation model and a 3D reconstruction model, including the following steps: Collecting a number of historical medical image data and corresponding historical patient diagnosis data, and preprocessing them to obtain a number of preprocessed historical medical image data and a number of preprocessed historical patient diagnosis data; Based on a number of pre-processed historical medical imaging data, a deep learning algorithm is used to build an image segmentation model and generate a number of historical segmentation images; Based on several historical segmentation images and corresponding pre-processed historical patient diagnosis data, a deep learning algorithm is used to construct a three-dimensional reconstruction model.

3. The three-dimensional modeling method of medical image data according to claim 2, characterized in that: The preprocessing of the historical medical image data includes data cleaning, Gaussian denoising, size normalization, image enhancement and labeling in sequence; The preprocessing of the historical patient diagnosis data includes data cleaning, Gaussian denoising and magnitude normalization performed in sequence.

4. The three-dimensional modeling method of medical image data according to claim 2, characterized in that: The three-dimensional reconstruction model is constructed based on the cGAN-MLP algorithm, and the three-dimensional reconstruction model includes a conditional information embedder and a conditional information processor both constructed based on the MLP algorithm, and a generator and a discriminator both constructed based on the RNN algorithm. The generator is connected to the conditional information embedder and the discriminator respectively, and the discriminator is connected to the conditional information processor.

5. The three-dimensional modeling method of medical image data according to claim 4, characterized in that: Based on several historical segmentation images and the corresponding pre-processed historical patient diagnosis data, a deep learning algorithm is used to construct a 3D reconstruction model, including the following steps: Setting corresponding real three-dimensional data for each historical segmentation image and corresponding pre-processed historical patient diagnosis data; Use the cGAN algorithm to build the initial generator and initial discriminator, and use the MLP algorithm to set the conditional information embedder for the initial generator and the conditional information processor for the initial discriminator; Integrate the initial generator, initial discriminator, conditional information embedder, and conditional information processor to obtain the initial 3D reconstruction model; Combine the first loss function of the initial generator and the second loss function of the initial discriminator to obtain a comprehensive loss function; Extracting historical image features of historical segmented images and historical data features of pre-processed historical patient diagnosis data; Use the conditional information embedder to conditionally embed historical image features and historical data features to obtain historical conditional information embedding features; Using the initial generator, embedding features based on historical condition information to generate three-dimensional data, thereby generating three-dimensional data; Using a condition information processor, condition information processing is performed based on the real three-dimensional data and the corresponding generated three-dimensional data to obtain historical condition information; Using the initial discriminator, data discrimination is performed based on the real 3D data, the corresponding generated 3D data, and the historical condition information to obtain the historical data discrimination result; Traverse all historical segmentation images and the corresponding pre-processed historical patient diagnosis data, repeat the above steps, and optimize the initial 3D reconstruction model; Use the comprehensive loss function to generate the historical loss value of the optimization training process. If the historical loss value is lower than the loss value threshold, the final 3D reconstruction model is output; otherwise, the optimization training continues.

6. The three-dimensional modeling method of medical image data according to claim 1, characterized in that: Collecting real-time medical image data and corresponding real-time patient diagnostic data, and using an image segmentation model to perform image segmentation on the real-time medical image data to obtain a real-time segmented image, includes the following steps: Collecting real-time medical imaging data and corresponding real-time patient diagnostic data, and inputting the real-time medical imaging data into an image segmentation model; Use the graph feature extraction module of the image segmentation model to extract several real-time feature graphs of real-time medical imaging data; Use the feature fusion module of the image segmentation model to fuse several real-time feature maps to obtain real-time fusion features; The image segmentation module of the image segmentation model is used to perform image segmentation on the real-time fusion features to obtain a real-time segmented image.

7. The three-dimensional modeling method of medical image data according to claim 5, characterized in that: Using the 3D reconstruction model, the real-time segmented image and the real-time patient diagnostic data are 3D reconstructed to obtain a real-time 3D model of the real-time medical imaging data, including the following steps: inputting the real-time segmentation image and the real-time patient diagnosis data into the three-dimensional reconstruction model, and extracting the real-time image features of the real-time segmentation image and the real-time data features of the real-time patient diagnosis data; Using the conditional information embedder of the 3D reconstruction model, conditionally embed the real-time image features and real-time data features to obtain real-time conditional information embedding features; Using a 3D reconstruction model generator, embed features based on real-time condition information to generate 3D data and obtain real-time 3D data; Based on the real-time three-dimensional data, three-dimensional reconstruction is performed to obtain a real-time three-dimensional model of the real-time medical imaging data.

8. A three-dimensional modeling system for medical imaging data, for implementing the three-dimensional modeling method according to any one of claims 1 to 7, characterized in that: The system comprises a model building unit, an image segmentation unit and a three-dimensional reconstruction unit which are connected in sequence.

Images