Segmentation method and device for magnetic resonance vascular structure imaging

Abstract

The invention relates to the technical field of medical images, in particular to a segmentation method and device for magnetic resonance vascular structure imaging. The method comprises the following steps: carrying out data preprocessing on a VAI image set; constructing a deep learning model, training the deep learning model by using the VAI image set after data preprocessing, and optimizing model parameters; and performing automatic tumor region segmentation on the new VAI image by using the trained and optimized deep learning model to generate a segmentation result. Through combination of deep learning models such as a convolutional neural network CNN and the like, features of a blood vessel structure can be automatically learned, and accurate segmentation is carried out for capillary changes of a tumor region.

Description

Technical Field

[0001] The present invention relates to the field of medical imaging technology, and in particular to a segmentation method and device for magnetic resonance vascular structure imaging. Background Art

[0002] Vascular structure imaging (VAI) is a technology based on magnetic resonance imaging (MRI) that can obtain quantitative information about vascular structure through different imaging sequences (such as gradient echo GE and spin echo SE). VAI can provide information about the size of blood vessels and microvessel density in tumors and surrounding tissues, which is crucial for evaluating tumor growth, staging, and treatment effects. Traditional VAI technology usually relies on manual or semi-automated segmentation methods and lacks accurate automatic segmentation of tumor areas.

[0003] Currently, existing brain tumor imaging methods include dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), magnetic resonance angiography (MRA), and functional MRI (fMRI), etc. These methods enhance the clarity of images by introducing contrast agents, especially in the assessment of angiogenesis and tumor blood flow.

[0004] However, most of the existing tumor segmentation methods are based on traditional image processing technology or deep learning models. Although traditional methods such as threshold segmentation, region growing and edge detection can achieve tumor region segmentation in some cases, their accuracy and robustness are poor. In recent years, the 3D U-Net model based on deep learning has been applied to the automatic segmentation task of medical images, which can better solve the segmentation problem of complex tumor morphology. However, the existing automatic segmentation methods often find it difficult to fully utilize the vascular information in VAI images to optimize the segmentation of tumor areas, and there are still limitations in the quantitative analysis of microvascular density around tumors.

[0005] In summary, the shortcomings of the prior art are:

[0006] 1. Insufficient segmentation accuracy: Although the existing automatic segmentation methods can handle the basic area segmentation of tumors, the segmentation accuracy is still low when the tumor boundary is blurred, the tumor is small, or the tumor is mixed with surrounding tissues.

[0007] 2. Insufficient consideration of vascular structure: Most existing segmentation methods do not fully consider the impact of vascular structure on the tumor area, especially the role of vascular features in VAI images in indicating tumor boundaries and microvascular density.

[0008] 3. Lack of quantitative analysis: Existing segmentation technologies mostly focus on qualitative description and lack precise quantitative analysis methods, and cannot provide a scientific basis for tumor treatment. Summary of the invention

[0009] The embodiment of the present invention provides a segmentation method and device for magnetic resonance vascular structure imaging, so as to at least solve the technical problem of low accuracy of existing tumor image segmentation.

[0010] According to an embodiment of the present invention, a segmentation method for magnetic resonance vascular structure imaging is provided, comprising the following steps:

[0011] S101: performing data preprocessing on the VAI image set;

[0012] S102: Build a deep learning model, use the VAI image set after data preprocessing to train the deep learning model, and optimize model parameters;

[0013] S103: Use the trained and optimized deep learning model to automatically segment the tumor area of ​​the new VAI image and generate a segmentation result.

[0014] Furthermore, the method further comprises:

[0015] S104: Perform quantitative analysis on the segmentation results.

[0016] Furthermore, step S104 specifically includes:

[0017] The vascular density and microvascular density indices of the tumor area are calculated based on the segmentation results, and a comprehensive evaluation is performed in combination with other imaging data in the VAI image set.

[0018] Furthermore, step S101 specifically includes:

[0019] The VAI image set is standardized to remove noise and enhance vascular structure information.

[0020] Furthermore, the vascular structure information includes vascular density, vascular size, and microvascular density.

[0021] Furthermore, step S102 specifically includes:

[0022] The 3D U-Net network architecture is used to train the model in combination with the vascular information in the VAI image.

[0023] Furthermore, high-quality labeled datasets are used in the training process to improve the accuracy of tumor segmentation by optimizing the loss function.

[0024] Furthermore, in step S101, the VAI image set includes the patient's VAI image and other related imaging data, and the other related imaging data includes T1-weighted imaging and SAGE imaging.

[0025] According to another embodiment of the present invention, a segmentation device for magnetic resonance vascular structure imaging is provided, comprising:

[0026] A preprocessing unit, used for performing data preprocessing on the VAI image set;

[0027] The training unit is used to build a deep learning model, train the deep learning model using the VAI image set after data preprocessing, and optimize the model parameters;

[0028] The segmentation unit is used to automatically segment the tumor area of ​​the new VAI image using the trained and optimized deep learning model to generate a segmentation result.

[0029] Furthermore, the device also includes:

[0030] The quantitative analysis unit is used to perform quantitative analysis on the segmentation results.

[0031] A storage medium stores a program file capable of implementing any one of the above-mentioned segmentation methods for magnetic resonance vascular structure imaging.

[0032] A processor is used to run a program, wherein when the program is run, any one of the above-mentioned magnetic resonance vascular structure imaging segmentation methods is executed.

[0033] The magnetic resonance vascular structure imaging segmentation method and device in the embodiment of the present invention, by combining deep learning models such as convolutional neural networks (CNN), can automatically learn the characteristics of vascular structures and accurately segment the microvascular changes in the tumor area. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0035] Figure 1 A flow chart of a segmentation method for magnetic resonance vascular structure imaging of the present invention;

[0036] Figure 2 A preferred flow chart of the segmentation method for magnetic resonance vascular structure imaging of the present invention;

[0037] Figure 3 A framework diagram of a segmentation method for magnetic resonance vascular structure imaging of the present invention;

[0038] Figure 4 It is a module diagram of the segmentation device of magnetic resonance vascular structure imaging of the present invention;

[0039] Figure 5 This is a preferred module diagram of the segmentation device for magnetic resonance vascular structure imaging of the present invention. DETAILED DESCRIPTION

[0040] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme 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 only 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 creative work should fall within the scope of protection of the present invention.

[0041] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0042] Example 1

[0043] According to an embodiment of the present invention, a segmentation method for magnetic resonance vascular structure imaging is provided. Figure 1 , including the following steps:

[0044] S101: performing data preprocessing on the VAI image set;

[0045] S102: Build a deep learning model, use the VAI image set after data preprocessing to train the deep learning model, and optimize model parameters;

[0046] S103: Use the trained and optimized deep learning model to automatically segment the tumor area of ​​the new VAI image and generate a segmentation result.

[0047] The magnetic resonance vascular structure imaging segmentation method in the embodiment of the present invention, by combining deep learning models such as convolutional neural networks (CNN), can automatically learn the characteristics of vascular structures and accurately segment the microvascular changes in the tumor area.

[0048] Among them, see Figure 2 , the method further comprises:

[0049] S104: Perform quantitative analysis on the segmentation results.

[0050] Existing tumor segmentation methods generally rely on traditional image processing technology or deep learning algorithms. When the tumor boundary is blurred and the image contrast is low, the tumor area may not be accurately segmented. In addition, the blood vessel density in the tumor area varies greatly, which increases the difficulty of segmentation, making it difficult for traditional segmentation algorithms to effectively distinguish between the tumor area and normal tissue. The technical problem to be solved by the present invention is how to effectively and accurately perform automatic segmentation of the tumor area based on vascular structure imaging (VAI) brain tumor image data, and evaluate key indicators such as the tumor's vascular structure and microvascular density through this segmentation result, so as to provide effective data support for early diagnosis, staging and treatment monitoring of tumors.

[0051] In view of the limitations of the above-mentioned traditional tumor segmentation methods, the present invention proposes a new tumor segmentation method based on magnetic resonance vascular imaging (VAI) technology. By combining deep learning models such as convolutional neural networks (CNN), it can automatically learn the characteristics of vascular structure and accurately segment the microvascular changes in the tumor area.

[0052] The purpose of the present invention is to provide an automatic tumor segmentation method based on VAI vascular structure imaging, which uses a deep learning algorithm and combines vascular structure information to accurately segment the tumor area and perform quantitative analysis, such as vascular density, microvascular density, etc., to provide reliable support for early diagnosis, staging, and treatment effect monitoring of tumors.

[0053] The basic contents of the technical solution of the present invention are as follows:

[0054] The present invention uses a deep learning algorithm based on VAI vascular structure imaging to automatically segment brain tumors. By using a convolutional neural network (CNN) or U-Net architecture, combined with vascular structure information in VAI images (such as vascular density, vascular size, microvascular density, etc.), accurate segmentation and quantitative analysis of tumor areas are achieved. The specific method includes:

[0055] 1. Data preprocessing: Standardize the VAI image, remove noise, and enhance the vascular structure information.

[0056] 2. Deep learning model training: Use VAI images and their annotated datasets to train deep learning models, optimize model parameters, and improve segmentation accuracy.

[0057] 3. Automatic tumor segmentation: Use the trained model to automatically segment the tumor area of ​​the new VAI image.

[0058] 4. Quantitative analysis: The segmentation results are used to calculate indicators such as vascular density and microvascular density in the tumor area, providing a basis for clinical monitoring of tumor growth and treatment.

[0059] The technical solution of the present invention is described in detail as follows:

[0060] The present invention conducted experiments on more than 100 brain tumor patients using 3T and 5T magnetic resonance equipment, and conducted animal experiments on glioma-bearing mice using 9.4T magnetic resonance equipment to verify the performance of magnetic resonance vascular structure imaging technology, and added U-Net to segment the tumor area to verify the accuracy and effectiveness of the technology.

[0061] See also Figure 3 The main technical contents of the present invention are as follows:

[0062] 1. Data acquisition and preprocessing: First, collect the patient's VAI images and other related imaging data (such as T1-weighted imaging, SAGE imaging, etc.). De-noise, standardize and register these data to ensure the alignment of different imaging data.

[0063] 2. Construction and training of deep learning models: Use the 3D U-Net network architecture and combine the vascular information in the VAI image to train the model. Use high-quality annotated data sets during training and optimize the loss function to improve the accuracy of tumor segmentation.

[0064] 3. Automatic segmentation and quantitative analysis: After the model training is completed, it is applied to new VAI image data to automatically segment the tumor area. Based on the segmentation results, further quantitative analysis is performed to calculate the characteristics of the tumor area, such as vascular density and microvascular density, and combined with other imaging data for comprehensive evaluation.

[0065] 4. Clinical application: Through precise segmentation and quantitative analysis of tumor areas, it helps doctors to make early diagnosis, stage tumors, monitor treatment effects, and provide a scientific basis for the formulation of personalized treatment plans.

[0066] Through the above technical scheme, the present invention can provide a more scientific, non-invasive and efficient technical means for the early diagnosis, treatment effect monitoring and efficacy evaluation of brain tumors, thereby improving the accuracy of tumor treatment and reducing the risks to patients.

[0067] The present invention has accurate tumor segmentation: by combining VAI imaging with deep learning algorithms, the present invention can extract detailed tumor and microvascular features from vascular structure imaging and perform accurate tumor segmentation.

[0068] The present invention has precise quantitative analysis: compared with the existing DCE-MRI and MRA technologies, the present invention can provide more accurate quantitative data, such as important indicators such as microvascular density and vascular size (such as vascular radius) in the tumor area, surpassing the limitations of the existing technology that mainly focuses on large vascular structures.

[0069] The present invention comprehensively evaluates the microvascular status: the present invention not only focuses on the angiogenesis in the tumor, but also can quantitatively evaluate the microvascular changes in the area surrounding the tumor. Through this comprehensive analysis, a more comprehensive basis can be provided for tumor staging, treatment planning and efficacy evaluation.

[0070] The key points and intended protection points of the present invention are:

[0071] 1. Application of VAI vascular structure information: This invention combines VAI vascular structure information with a deep learning model for the first time to achieve accurate segmentation of the tumor area.

[0072] 2. Automated tumor segmentation method: The present invention provides an automated tumor segmentation method, which avoids the tediousness and inaccuracy of manual segmentation.

[0073] 3. Quantitative analysis of vascular density and microvascular density: The present invention provides support for the diagnosis and treatment of tumors by quantitatively analyzing key indicators such as vascular density and microvascular density in the tumor area.

[0074] Compared with the prior art, the advantages of the present invention are:

[0075] 1. Combining vascular structure information: Most existing technologies focus on image-based tumor segmentation, ignoring the impact of vascular structure on tumor boundaries and microvascular density. The present invention can more accurately identify tumor areas by combining vascular information in VAI images.

[0076] 2. High degree of automation: Compared with traditional manual segmentation methods or semi-automatic segmentation methods, the automated tumor segmentation method of the present invention is more efficient and accurate, reducing human errors.

[0077] 3. Strong quantitative analysis capability: Existing technologies are mostly limited to qualitative analysis, while the present invention can provide accurate quantitative data, such as vascular density, microvascular density, etc., to help doctors make more scientific treatment decisions.

[0078] The present invention has been proven to be feasible through experiments, simulations, and use. The present invention has been verified to be feasible through simulation experiments. Through training and testing on multiple tumor data sets, the experimental results show that the method can achieve high-precision segmentation in different types of tumors and can successfully extract quantitative indicators such as blood vessel density and microvessel density in the tumor area.

[0079] The altered design (alternative scheme) and other uses of the present invention are as follows: the deep learning model of the present invention can be adjusted according to different tumor types to adapt to the segmentation tasks of other types of tumors (such as lung cancer, breast cancer, etc.). In addition to brain tumor segmentation, the technology of the present invention can also be widely used in other automatic segmentation tasks of medical images, such as microvascular assessment of cardiovascular disease, diabetes, etc., and has broad application prospects.

[0080] Example 2

[0081] According to another embodiment of the present invention, a segmentation device for magnetic resonance vascular structure imaging is provided. Figure 4 ,include:

[0082] A preprocessing unit 201, used for performing data preprocessing on a VAI image set;

[0083] A training unit 202 is used to build a deep learning model, train the deep learning model using the VAI image set after data preprocessing, and optimize model parameters;

[0084] The segmentation unit 203 is used to automatically segment the tumor area of ​​the new VAI image using the trained and optimized deep learning model to generate a segmentation result.

[0085] The segmentation device for magnetic resonance vascular structure imaging in the embodiment of the present invention, by combining deep learning models such as convolutional neural networks (CNN), can automatically learn the characteristics of vascular structures and accurately segment the microvascular changes in the tumor area.

[0086] Among them, see Figure 5 , the device further comprises:

[0087] The quantitative analysis unit 204 is used to perform quantitative analysis on the segmentation result.

[0088] Existing tumor segmentation methods generally rely on traditional image processing technology or deep learning algorithms. When the tumor boundary is blurred and the image contrast is low, the tumor area may not be accurately segmented. In addition, the blood vessel density in the tumor area varies greatly, which increases the difficulty of segmentation, making it difficult for traditional segmentation algorithms to effectively distinguish between the tumor area and normal tissue. The technical problem to be solved by the present invention is how to effectively and accurately perform automatic segmentation of the tumor area based on vascular structure imaging (VAI) brain tumor image data, and evaluate key indicators such as the tumor's vascular structure and microvascular density through this segmentation result, so as to provide effective data support for early diagnosis, staging and treatment monitoring of tumors.

[0089] In view of the limitations of the above-mentioned traditional tumor segmentation methods, the present invention proposes a new tumor segmentation method based on magnetic resonance vascular imaging (VAI) technology. By combining deep learning models such as convolutional neural networks (CNN), it can automatically learn the characteristics of vascular structure and accurately segment the microvascular changes in the tumor area.

[0090] The purpose of the present invention is to provide an automatic tumor segmentation device based on VAI vascular structure imaging, which uses a deep learning algorithm and combines vascular structure information to accurately segment the tumor area and perform quantitative analysis, such as vascular density, microvascular density, etc., to provide reliable support for early diagnosis, staging, and treatment effect monitoring of tumors.

[0091] The basic contents of the technical solution of the present invention are as follows:

[0092] The present invention uses a deep learning algorithm based on VAI vascular structure imaging to automatically segment brain tumors. By using a convolutional neural network (CNN) or U-Net architecture, combined with vascular structure information in VAI images (such as vascular density, vascular size, microvascular density, etc.), accurate segmentation and quantitative analysis of tumor areas are achieved. The specific method includes:

[0093] 1. Data preprocessing: Standardize the VAI images, remove noise, and enhance vascular structure information.

[0094] 2. Deep learning model training: Use VAI images and their annotated datasets to train deep learning models, optimize model parameters, and improve segmentation accuracy.

[0095] 3. Automatic tumor segmentation: Use the trained model to automatically segment the tumor area of ​​the new VAI image.

[0096] 4. Quantitative analysis: Calculate indicators such as vascular density and microvascular density in the tumor area through segmentation results, providing a basis for clinical monitoring of tumor growth and treatment.

[0097] The technical solution of the present invention is described in detail as follows:

[0098] The present invention conducted experiments on more than 100 brain tumor patients using 3T and 5T magnetic resonance equipment, and conducted animal experiments on glioma-bearing mice using 9.4T magnetic resonance equipment to verify the performance of magnetic resonance vascular structure imaging technology, and added U-Net to segment the tumor area to verify the accuracy and effectiveness of the technology.

[0099] See also Figure 3 The main technical contents of the present invention are as follows:

[0100] 1. Data acquisition and preprocessing: First, collect the patient's VAI images and other related imaging data (such as T1-weighted imaging, SAGE imaging, etc.). De-noise, standardize and register these data to ensure the alignment of different imaging data.

[0101] 2. Construction and training of deep learning models: Use the 3D U-Net network architecture and combine the vascular information in the VAI image to train the model. Use high-quality annotated data sets during training and optimize the loss function to improve the accuracy of tumor segmentation.

[0102] 3. Automatic segmentation and quantitative analysis: After the model training is completed, it is applied to new VAI image data to automatically segment the tumor area. Based on the segmentation results, further quantitative analysis is performed to calculate the characteristics of the tumor area, such as vascular density and microvascular density, and combined with other imaging data for comprehensive evaluation.

[0103] 4. Clinical application: Through precise segmentation and quantitative analysis of tumor areas, it helps doctors to make early diagnosis, stage tumors, monitor treatment effects, and provide a scientific basis for the formulation of personalized treatment plans.

[0104] Through the above technical scheme, the present invention can provide a more scientific, non-invasive and efficient technical means for the early diagnosis, treatment effect monitoring and efficacy evaluation of brain tumors, thereby improving the accuracy of tumor treatment and reducing the risks to patients.

[0105] The present invention has accurate tumor segmentation: by combining VAI imaging with deep learning algorithms, the present invention can extract detailed tumor and microvascular features from vascular structure imaging and perform accurate tumor segmentation.

[0106] The present invention has precise quantitative analysis: compared with the existing DCE-MRI and MRA technologies, the present invention can provide more accurate quantitative data, such as important indicators such as microvascular density and vascular size (such as vascular radius) in the tumor area, surpassing the limitations of the existing technology that mainly focuses on large vascular structures.

[0107] The present invention comprehensively evaluates the microvascular status: the present invention not only focuses on the angiogenesis in the tumor, but also can quantitatively evaluate the microvascular changes in the area surrounding the tumor. Through this comprehensive analysis, a more comprehensive basis can be provided for tumor staging, treatment planning and efficacy evaluation.

[0108] The key points and intended protection points of the present invention are:

[0109] 1. Application of VAI vascular structure information: This invention combines VAI vascular structure information with a deep learning model for the first time to achieve accurate segmentation of the tumor area.

[0110] 2. Automated tumor segmentation method: The present invention provides an automated tumor segmentation method, which avoids the tediousness and inaccuracy of manual segmentation.

[0111] 3. Quantitative analysis of vascular density and microvascular density: The present invention provides support for the diagnosis and treatment of tumors by quantitatively analyzing key indicators such as vascular density and microvascular density in the tumor area.

[0112] Compared with the prior art, the advantages of the present invention are:

[0113] 1. Combining vascular structure information: Most existing technologies focus on image-based tumor segmentation, ignoring the impact of vascular structure on tumor boundaries and microvascular density. The present invention can more accurately identify tumor areas by combining vascular information in VAI images.

[0114] 2. High degree of automation: Compared with traditional manual segmentation methods or semi-automatic segmentation methods, the automated tumor segmentation method of the present invention is more efficient and accurate, reducing human errors.

[0115] 3. Strong quantitative analysis capability: Existing technologies are mostly limited to qualitative analysis, while the present invention can provide accurate quantitative data, such as vascular density, microvascular density, etc., to help doctors make more scientific treatment decisions.

[0116] The present invention has been proven to be feasible through experiments, simulations, and use. The present invention has been verified to be feasible through simulation experiments. Through training and testing on multiple tumor data sets, the experimental results show that the method can achieve high-precision segmentation in different types of tumors and can successfully extract quantitative indicators such as blood vessel density and microvessel density in the tumor area.

[0117] The altered design (alternative scheme) and other uses of the present invention are as follows: the deep learning model of the present invention can be adjusted according to different tumor types to adapt to the segmentation tasks of other types of tumors (such as lung cancer, breast cancer, etc.). In addition to brain tumor segmentation, the technology of the present invention can also be widely used in other automatic segmentation tasks of medical images, such as microvascular assessment of cardiovascular disease, diabetes, etc., and has broad application prospects.

[0118] Example 3

[0119] A storage medium stores a program file capable of implementing any one of the above-mentioned segmentation methods for magnetic resonance vascular structure imaging.

[0120] Example 4

[0121] A processor is used to run a program, wherein when the program is run, any one of the above-mentioned magnetic resonance vascular structure imaging segmentation methods is executed.

[0122] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0123] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0124] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the system embodiments described above are only schematic. For example, the division of units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0125] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed over multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0126] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0127] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0128] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A segmentation method for magnetic resonance vascular structure imaging, characterized in that: The following steps are involved: S101: performing data preprocessing on the VAI image set; S102: Build a deep learning model, use the VAI image set after data preprocessing to train the deep learning model, and optimize model parameters; S103: Use the trained and optimized deep learning model to automatically segment the tumor area of ​​the new VAI image and generate a segmentation result.

2. The segmentation method for magnetic resonance vascular structure imaging according to claim 1, characterized in that: The method further comprises: S104: Perform quantitative analysis on the segmentation results.

3. The segmentation method for magnetic resonance vascular structure imaging according to claim 2, characterized in that: Step S104 specifically includes: The vascular density and microvascular density indices of the tumor area are calculated based on the segmentation results, and a comprehensive evaluation is performed in combination with other imaging data in the VAI image set.

4. The segmentation method for magnetic resonance vascular structure imaging according to claim 1, characterized in that: Step S101 specifically includes: The VAI image set is standardized to remove noise and enhance vascular structure information.

5. The segmentation method for magnetic resonance vascular structure imaging according to claim 4, characterized in that: Vascular structure information includes vascular density, vascular size, and microvessel density.

6. The segmentation method for magnetic resonance vascular structure imaging according to claim 1, characterized in that: Step S102 specifically includes: The 3D U-Net network architecture is used to train the model in combination with the vascular information in the VAI image.

7. The segmentation method for magnetic resonance vascular structure imaging according to claim 6, characterized in that: High-quality labeled datasets are used during the training process to improve the accuracy of tumor segmentation by optimizing the loss function.

8. The segmentation method for magnetic resonance vascular structure imaging according to claim 1, characterized in that: In step S101, the VAI image set includes the patient's VAI image and other related imaging data, and the other related imaging data includes T1-weighted imaging and SAGE imaging.

9. A segmentation device for magnetic resonance vascular structure imaging, characterized in that: include: A preprocessing unit, used for performing data preprocessing on the VAI image set; The training unit is used to build a deep learning model, train the deep learning model using the VAI image set after data preprocessing, and optimize the model parameters; The segmentation unit is used to automatically segment the tumor area of ​​the new VAI image using the trained and optimized deep learning model to generate a segmentation result.

10. The segmentation device for magnetic resonance vascular structure imaging according to claim 9, characterized in that: The device also includes: The quantitative analysis unit is used to perform quantitative analysis on the segmentation results.

Images