Blood vessel risk assessment method, computer device and storage medium

Abstract

The present application relates to a blood vessel risk assessment method, computer equipment and storage medium. The method comprises: inputting a medical image into a preset segmentation model to obtain a mask image of an artery of interest in the medical image; determining a preset tissue region where the artery of interest is located based on the mask image of the artery of interest; extracting image features of the preset tissue region where the artery of interest is located to obtain imaging features of the preset tissue region where the artery of interest is located; and obtaining a risk assessment result of the artery of interest based on the imaging features and a preset classification model. The use of this method can improve the efficiency of the risk assessment result of the artery of interest.

Description

Technical Field

[0001] The present application relates to the technical field of medical image processing, and in particular to a blood vessel risk assessment method, computer equipment and storage medium. Background Art

[0002] Most arterial diseases are caused by unstable, non-obstructive atherosclerotic plaques. The formation and rupture of atherosclerotic plaques often affect the tissue surrounding the blood vessels. Studies have shown that the characteristics of atherosclerotic plaques are related to the tissue surrounding the arteries. Therefore, the analysis of the tissue surrounding the arteries helps to effectively evaluate the characteristics of atherosclerotic plaques that cause arterial disease.

[0003] In traditional technology, the periarterial fat area is usually manually outlined, and then the fat attenuation index of the outlined periarterial fat area is obtained, and the characteristics of atherosclerotic plaques are evaluated based on the periarterial fat attenuation index.

[0004] However, traditional evaluation methods have the problem of low evaluation efficiency. Summary of the invention

[0005] Based on this, it is necessary to provide a vascular risk assessment method, computer device and storage medium that can improve the efficiency of assessing the characteristics of atherosclerotic plaques that cause arterial diseases in order to address the above technical problems.

[0006] A method for assessing blood vessel risk, the method comprising:

[0007] Inputting the medical image into a preset segmentation model to obtain a mask image of an artery of interest in the medical image;

[0008] Determining a preset tissue region where the artery of interest is located according to the mask image of the artery of interest;

[0009] Extracting image features of a preset tissue region where the artery of interest is located, and obtaining imaging omics features of the preset tissue region where the artery of interest is located;

[0010] According to the radiomics features and the preset classification model, a risk assessment result of the artery of interest is obtained.

[0011] In one embodiment, determining the preset tissue region where the artery of interest is located according to the mask image of the artery of interest includes:

[0012] Obtaining the lumen diameter of the arterial vessel of interest according to the mask image of the arterial vessel of interest;

[0013] The lumen diameter of the artery of interest is enlarged, and the area within the enlarged lumen diameter is determined as the preset tissue area where the artery of interest is located.

[0014] In one embodiment, obtaining the lumen diameter of the arterial vessel of interest according to the mask image of the arterial vessel of interest includes:

[0015] Extracting the center line of the arterial vessel of interest from the mask image of the arterial vessel of interest;

[0016] For each point on the centerline of the artery of interest, determine the cross section where each point is located;

[0017] The contour of the arterial vessel of interest is extracted based on the cross section, and the lumen diameter of the arterial vessel of interest is obtained according to the contour of the arterial vessel of interest.

[0018] In one embodiment, the method further comprises:

[0019] Acquire a feature map obtained during the process of segmenting the artery of interest in the medical image by the segmentation model;

[0020] Obtaining a target feature according to the feature map and the radiomics feature;

[0021] The target feature is input into the classification model to obtain a risk assessment result of the artery of interest.

[0022] In one embodiment, obtaining a target feature according to the feature map and the radiomics feature includes:

[0023] Performing feature conversion on the feature map to obtain converted features; the dimension of the converted features is the same as the dimension of the radiomics features;

[0024] The converted features and the radiomics features are fused, and the fused features are determined as the target features.

[0025] In one embodiment, determining the fused feature as the target feature includes:

[0026] The fused features are extracted using a preset feature extraction algorithm to extract nonlinearly related features from the fused features, and the extracted features are determined as the target features.

[0027] In one embodiment, the preset feature extraction algorithm includes principal component analysis.

[0028] In one embodiment, the preset classification model is a model constructed based on a machine learning algorithm, or a model constructed based on a neural network model including a fully connected layer; the machine learning algorithm includes a random forest method, a support vector machine, a k-nearest neighbor method, a decision tree, a logistic regression model, an XGboost algorithm, a quadratic discriminant analysis method, a stochastic gradient descent method, and an Adaboost algorithm.

[0029] A blood vessel risk assessment device, comprising:

[0030] A first acquisition module is used to input the medical image into a preset segmentation model to obtain a mask image of an artery of interest in the medical image;

[0031] A determination module, configured to determine a preset tissue region where the artery of interest is located according to the mask image of the artery of interest;

[0032] An extraction module, used to extract image features of a preset tissue region where the artery of interest is located, and obtain image omics features of the preset tissue region where the artery of interest is located;

[0033] The second acquisition module is used to obtain the risk assessment result of the artery of interest according to the imaging genomics characteristics and a preset classification model.

[0034] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0035] Inputting the medical image into a preset segmentation model to obtain a mask image of an artery of interest in the medical image;

[0036] Determining a preset tissue region where the artery of interest is located according to the mask image of the artery of interest;

[0037] Extracting image features of a preset tissue region where the artery of interest is located, and obtaining imaging omics features of the preset tissue region where the artery of interest is located;

[0038] According to the radiomics features and the preset classification model, a risk assessment result of the artery of interest is obtained.

[0039] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0040] Inputting the medical image into a preset segmentation model to obtain a mask image of an artery of interest in the medical image;

[0041] Determining a preset tissue region where the artery of interest is located according to the mask image of the artery of interest;

[0042] Extracting image features of a preset tissue region where the artery of interest is located, and obtaining imaging omics features of the preset tissue region where the artery of interest is located;

[0043] According to the radiomics features and the preset classification model, a risk assessment result of the artery of interest is obtained.

[0044] The above-mentioned vascular risk assessment method, device, computer equipment and storage medium can quickly obtain a mask image of the artery of interest in the medical image by inputting the medical image into a preset segmentation model, so that the preset tissue area where the artery of interest is located can be quickly determined based on the mask image of the artery of interest, and the image features of the preset tissue area where the artery of interest is located can be extracted, thereby improving the efficiency of obtaining the imaging genomics features of the preset tissue area where the artery of interest is located, and further improving the efficiency of assessing the risk of the artery of interest based on the extracted imaging genomics features and the preset classification model to obtain the risk assessment result of the artery of interest; in addition, the risk of the artery of interest is assessed based on the imaging genomics features of the preset tissue area where the artery of interest is located and the preset classification model, thereby avoiding the problem of inaccurate assessment using a single feature and improving the accuracy of the risk assessment result of the artery of interest. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A diagram showing an application environment of a blood vessel risk assessment method in one embodiment;

[0046] Figure 2 A schematic diagram of a flow chart of a blood vessel risk assessment method in one embodiment;

[0047] Figure 2a A schematic diagram of the structure of a segmentation model preset in one embodiment;

[0048] Figure 3 is a schematic flow chart of a blood vessel risk assessment method in another embodiment;

[0049] Figure 4 is a schematic flow chart of a blood vessel risk assessment method in another embodiment;

[0050] Figure 5 is a schematic flow chart of a blood vessel risk assessment method in another embodiment;

[0051] Figure 6 is a schematic flow chart of a blood vessel risk assessment method in another embodiment;

[0052] Figure 7 FIG. 4 is a structural block diagram of a blood vessel risk assessment device in one embodiment. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0054] The blood vessel risk assessment method provided in the embodiment of the present application can be applied to Figure 1 The computer device shown. The computer device includes a processor and a memory connected by a system bus, and a computer program is stored in the memory. When the processor executes the computer program, the steps of the following method embodiment can be executed. Optionally, the computer device may also include a network interface, a display screen and an input device. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory, and the non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. Optionally, the computer device can be a server, a personal computer, a personal digital assistant, or other terminal devices, such as a tablet computer, a mobile phone, etc., or a cloud or remote server. The embodiment of the present application does not limit the specific form of the computer device.

[0055] In one embodiment, Figure 2 As shown, a method for risk assessment of blood vessels is provided, and the method is applied to Figure 1 The computer device in the example is used to illustrate, including the following steps:

[0056] S201, inputting a medical image into a preset segmentation model to obtain a mask image of an artery of interest in the medical image.

[0057] The medical image may be a coronary artery image of the subject or a head and neck artery image of the subject, and accordingly, the artery of interest may be a coronary artery or a head and neck artery. Optionally, the medical image may be a computed tomography (CT) image or a magnetic resonance imaging (MRI) image.

[0058] Specifically, the computer device can input the medical image into a preset segmentation model, extract the features of the medical image through the segmentation model, and use the features of the medical image to segment the arteries of interest in the medical image to obtain a mask image of the arteries of interest in the medical image. Optionally, the preset segmentation model can be composed of an input layer, several convolution modules, several pooling modules, several deconvolution modules and an output module, and the network structure of the segmentation model can be Figure 2a The V-shaped structure shown or a similar structure.

[0059] S202: Determine a preset tissue region where the artery of interest is located according to the mask image of the artery of interest.

[0060] The preset tissue region where the artery of interest is located may be the region where the mask image of the artery of interest is located, or may be a region obtained by expanding the region where the mask image of the artery of interest is located. Optionally, the computer device may determine the length of the mask image of the artery of interest, and then expand the length of the mask image of the artery of interest, and determine the image region after the expansion as the preset tissue region where the artery of interest is located.

[0061] S203, extracting image features of a preset tissue region where the artery of interest is located, and obtaining imaging omics features of the preset tissue region where the artery of interest is located.

[0062] Optionally, in this embodiment, the computer device can extract the image features of the preset tissue region where the artery of interest is located through a preset feature extraction algorithm to obtain the imaging features of the preset tissue region where the artery of interest is located, or the computer device can calculate the grayscale value of the preset tissue region where the artery of interest is located to obtain the grayscale features of the preset tissue region where the artery of interest is located, thereby obtaining the imaging features of the preset tissue region where the artery of interest is located. Optionally, the imaging features of the preset tissue region include, but are not limited to, shape features, grayscale statistics features, texture features, and high-order features. Among them, shape features include three-dimensional features such as volume, surface area, and similarity to a ball; grayscale statistics features include the average value, peak value, maximum value, minimum value, etc. of the image voxel intensity; texture features include the features of the grayscale distribution of voxels and their surrounding spatial neighborhoods calculated by grayscale co-occurrence matrix, grayscale run matrix, etc.; high-order features include grayscale statistics features and texture features extracted after filtering the image using mean filtering, Gaussian filtering, wavelet transform, etc.

[0063] S203, obtaining risk assessment results of the arteries of interest based on the radiomics features and the preset classification model.

[0064] Specifically, the computer device can input the imaging omics features of the preset tissue area where the artery of interest is located into the preset classification model to obtain the risk assessment result of the artery of interest. It should be noted that in this embodiment, the risk assessment result of the artery of interest is used to characterize the characteristic assessment result of the plaque in the artery of interest, that is, to characterize whether there is a plaque in the artery of interest, whether the plaque in the artery of interest is ruptured, etc., and this embodiment is not limited here.

[0065] In the above-mentioned vascular risk assessment method, by inputting the medical image into a preset segmentation model, the mask image of the artery of interest in the medical image can be quickly obtained, so that the preset tissue region where the artery of interest is located can be quickly determined based on the mask image of the artery of interest, and the image features of the preset tissue region where the artery of interest is located can be extracted, thereby improving the efficiency of obtaining the imaging genomics features of the preset tissue region where the artery of interest is located, thereby improving the efficiency of assessing the risk of the artery of interest based on the extracted imaging genomics features and the preset classification model, and obtaining the risk assessment results of the artery of interest; in addition, the risk of the artery of interest is assessed based on the imaging genomics features of the preset tissue region where the artery of interest is located and the preset classification model, thereby avoiding the problem of inaccurate assessment using a single feature and improving the accuracy of the risk assessment results of the artery of interest.

[0066] Furthermore, in the above scenario where the preset tissue region where the artery of interest is located is determined based on the mask image of the artery of interest, the preset tissue region where the artery of interest is located can be determined based on the lumen diameter of the artery of interest. Figure 3 As shown, the above S202 includes:

[0067] S301, obtaining the lumen diameter of the arterial vessel of interest according to the mask image of the arterial vessel of interest.

[0068] Among them, the artery of interest can be regarded as a cylinder, and the lumen diameter of the artery of interest can be understood as the diameter of the cylinder. Optionally, the computer device can use a morphological feature extraction method to extract the center line of the artery of interest from the mask image of the artery of interest, determine the cross section where the point on the center line of the artery of interest is located, extract the contour of the artery of interest based on the cross section where each point is located, and obtain the lumen diameter of the artery of interest according to the contour of the artery of interest. Further, optionally, the computer device can calculate the major and minor axes of the contour of the artery of interest, and obtain the lumen diameter of the artery of interest according to the major and minor axes of the contour of the artery of interest.

[0069] S302, enlarging the lumen diameter of the artery of interest, and determining the area within the enlarged lumen diameter as the preset tissue area where the artery of interest is located.

[0070] Specifically, the computer device can use a preset expansion coefficient to expand the lumen diameter of the artery of interest, and determine the area within the expanded lumen diameter as the preset tissue area where the artery of interest is located. It should be noted that the preset tissue area where the artery of interest is located may include the artery of interest, or may not include the artery of interest. Optionally, the preset expansion coefficient may be to expand the lumen diameter of the artery of interest by one time, or to expand the lumen diameter of the artery of interest by two times. Optionally, the computer device can also calculate the distance field from the area outside the artery of interest to the mask image of the artery of interest for the mask image of the artery of interest, and determine the distance of the preset multiple of the lumen diameter of the artery of interest obtained above as the target distance, and determine whether the obtained target distance is less than each distance value in the distance field. If so, the preset tissue area where the artery of interest is located is determined according to the target distance. For example, the area within the target distance range can be determined as the preset tissue area where the artery of interest is located.

[0071] In this embodiment, based on the mask image of the artery of interest, the lumen diameter of the artery of interest can be accurately obtained. Since the accuracy of the lumen diameter of the artery of interest is improved, the accuracy of the lumen diameter of the expanded artery of interest is ensured, and the area within the expanded lumen diameter can be accurately determined as the preset tissue area where the artery of interest is located, thereby improving the accuracy of the preset tissue area where the artery of interest is located.

[0072] In the above scenario where the medical image is input into the preset segmentation model, the computer device can also obtain a feature map of the artery of interest in the medical image, and combine the feature map with the above imaging features to obtain a risk assessment result of the artery of interest. Figure 4 As shown, the above method also includes:

[0073] S401, obtaining a feature map obtained by a segmentation model during segmentation of an artery of interest in a medical image.

[0074] Specifically, in this embodiment, the computer device obtains a feature map obtained during the segmentation of the artery of interest in the medical image by the segmentation model. It can be understood that after the computer device inputs the medical image into the preset segmentation model, the segmentation model will perform a downsampling operation on the medical image, and the obtained feature map can be a feature map in any downsampling operation process, or a feature map corresponding to the first upsampling.

[0075] S402, obtaining target features according to the feature graph and the radiomics features.

[0076] Specifically, the computer device can obtain the target feature according to the feature map of the artery of interest and the fusion feature of the above-mentioned imaging omics feature. Optionally, the computer device can convert the feature map of the artery of interest to obtain the converted feature, perform feature fusion on the converted feature and the imaging omics feature, and determine the fused feature as the above-mentioned target feature, wherein the dimension of the converted feature is the same as the dimension of the above-mentioned imaging omics feature. It can be understood that only features with the same dimension can be feature fused. For example, if the imaging omics feature is a one-dimensional feature, the computer device can convert the feature map of the artery of interest into a one-dimensional feature. If the imaging omics feature is a two-dimensional feature, the computer device can convert the feature map of the artery of interest into a two-dimensional feature. It can be understood that after the converted feature and the imaging omics feature are feature fused, there may be repeated features in the fused feature. Further, the computer device can extract the features of the fused feature using a preset feature extraction algorithm, extract the nonlinearly related features in the fused feature, and determine the extracted feature as the target feature. For example, the computer device can extract the features of the fused feature using the principal component analysis method. Optionally, the preset feature extraction algorithm may also include, but is not limited to, using analysis of variance (ANOVA) to perform univariate selection on features, using the least absolute shrinkage and selection operator (LASSO) algorithm to perform multivariate selection, etc.

[0077] S403, inputting the target features into a classification model to obtain a risk assessment result of the artery of interest.

[0078] Specifically, after the computer device obtains the above-mentioned target features, the target features include both features corresponding to the feature map of the artery of interest and the imaging features of the preset tissue area where the artery of interest is located. The computer device can input the target features into the above-mentioned preset classification model, and use the features corresponding to the feature map of the artery of interest and the imaging features of the preset tissue area where the artery of interest is located to obtain the risk assessment results of the above-mentioned artery of interest.

[0079] In this embodiment, the computer device obtains a feature map obtained during the segmentation process of the artery of interest in the medical image by acquiring a segmentation model, and obtains a target feature based on the obtained feature map and the imaging features of the preset tissue region where the artery of interest is located. Since the target feature includes both the features corresponding to the feature map of the artery of interest and the imaging features of the preset tissue region where the artery of interest is located, the richness and accuracy of the obtained target feature are improved, so that the obtained target feature can be input into a preset classification model to accurately obtain the risk assessment result of the artery of interest, thereby improving the accuracy of the obtained risk assessment result of the artery of interest.

[0080] In the above-mentioned scenario where the imaging features of the preset tissue area where the artery of interest is located are input into the preset classification model, or in the scenario where the above-mentioned target features are input into the preset classification model, in one embodiment, the above-mentioned preset classification model is a model constructed based on a machine learning algorithm, or a model constructed based on a neural network model including a fully connected layer; the above-mentioned machine learning algorithms include random forest method, support vector machine, k-nearest neighbor method, decision tree, logistic regression model, XGboost algorithm, quadratic discriminant analysis method, stochastic gradient descent method, and Adaboost algorithm.

[0081] Specifically, the computer device can use a machine learning algorithm to construct the above-mentioned preset classification model to obtain a model constructed based on the machine learning algorithm, or use a neural network model including a fully connected layer to construct the above-mentioned preset classification model. It should be noted here that if the preset classification model is a model constructed based on a neural network model including a fully connected layer, the preset classification model does not include a convolutional layer. Among them, the machine learning algorithm in the model constructed based on the machine learning algorithm includes a random forest method, a support vector machine, a k-nearest neighbor method, a decision tree, a logistic regression model, an XGboost algorithm, a quadratic discriminant analysis method, a stochastic gradient descent method, and an Adaboost algorithm.

[0082] In this embodiment, the preset classification model is a model built based on a machine learning algorithm, or a model built based on a neural network model including a fully connected layer, which expands the usage scenarios of the classification model and enables the classification model to meet more usage scenarios.

[0083] To facilitate understanding by those skilled in the art, the following is a detailed description of the blood vessel risk assessment method provided in this application. Figure 5 and Figure 6 , the method may include:

[0084] S1, inputting the medical image into a preset segmentation model to obtain a mask image of the artery of interest in the medical image and a feature map of the medical image; the feature map of the medical image is a feature map obtained during the segmentation model segmenting the artery of interest in the medical image.

[0085] S2, extracting the center line of the artery of interest from the mask image of the artery of interest by using a morphological feature extraction method.

[0086] S3, for each point on the center line of the artery of interest, determine the cross section where each point is located.

[0087] S4, extracting the contour of the artery of interest based on the cross section, and obtaining the lumen diameter of the artery of interest according to the contour of the artery of interest.

[0088] S5, enlarging the lumen diameter of the artery of interest, and determining the area within the enlarged lumen diameter as the preset tissue area where the artery of interest is located.

[0089] S6, extracting the image features of the preset tissue region where the artery of interest is located, and obtaining the imaging omics features of the preset tissue region where the artery of interest is located.

[0090] S7, performing feature transformation on the feature map to obtain transformed features; the dimension of the transformed features is the same as the dimension of the imaging omics features.

[0091] S8, performing feature fusion on the converted features and the radiomics features to obtain fused features.

[0092] S9, extracting features from the fused features using a preset feature extraction algorithm, extracting nonlinearly related features from the fused features, and determining the extracted features as target features; the preset feature extraction algorithm includes a principal component analysis method.

[0093] S10, inputting the target features into the classification model to obtain the risk assessment results of the arteries of interest; the preset classification model is a model constructed based on a machine learning algorithm, or a model constructed based on a neural network model including a fully connected layer; the machine learning algorithms include random forest method, support vector machine, k-nearest neighbor method, decision tree, logistic regression model, XGboost algorithm, quadratic discriminant analysis method, stochastic gradient descent method, and Adaboost algorithm.

[0094] It should be noted that for the description in the above S1-S10, reference may be made to the relevant description in the above embodiment, and the effects are similar, which will not be repeated in this embodiment.

[0095] It should be understood that although Figures 2 to 6The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figures 2 to 6 At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0096] In one embodiment, Figure 7 As shown, a blood vessel risk assessment device is provided, comprising: a first acquisition module, a determination module, an extraction module, and a second acquisition module, wherein:

[0097] The first acquisition module is used to input the medical image into a preset segmentation model to obtain a mask image of the artery of interest in the medical image.

[0098] The determination module is used to determine the preset tissue area where the artery of interest is located according to the mask image of the artery of interest.

[0099] The extraction module is used to extract the image features of the preset tissue area where the artery of interest is located, and obtain the imaging features of the preset tissue area where the artery of interest is located.

[0100] The second acquisition module is used to obtain the risk assessment results of the arteries of interest based on the imaging genomics features and the preset classification model.

[0101] The blood vessel risk assessment device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0102] Based on the above embodiment, optionally, the above determination module includes: a first acquisition unit and a determination unit, wherein:

[0103] The first acquisition unit is used to obtain the lumen diameter of the arterial blood vessel of interest according to the mask image of the arterial blood vessel of interest.

[0104] The determination unit is used to expand the lumen diameter of the artery of interest and determine the area within the expanded lumen diameter as the preset tissue area where the artery of interest is located.

[0105] The blood vessel risk assessment device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0106] Based on the above embodiment, optionally, the above first acquisition unit is used to extract the center line of the artery of interest from the mask image of the artery of interest; determine the cross-section of each point on the center line of the artery of interest; extract the contour of the artery of interest based on the cross-section, and obtain the lumen diameter of the artery of interest according to the contour of the artery of interest.

[0107] The blood vessel risk assessment device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0108] On the basis of the above embodiment, optionally, the above device further includes: a third acquisition module, a fourth acquisition module and a fifth acquisition module, wherein:

[0109] The third acquisition module is used to obtain the feature map obtained by the segmentation model during the segmentation of the artery of interest in the medical image.

[0110] The fourth acquisition module is used to obtain target features according to the feature map and the imaging omics features.

[0111] The fifth acquisition module is used to input the target features into the classification model to obtain the risk assessment results of the arteries of interest.

[0112] The blood vessel risk assessment device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0113] On the basis of the above embodiment, optionally, the fourth acquisition module includes: a second acquisition unit and a fusion unit, wherein:

[0114] The second acquisition unit is used to perform feature conversion on the feature map to obtain converted features; the dimension of the converted features is the same as the dimension of the imaging omics features.

[0115] The fusion unit is used to fuse the converted features and the radiomics features, and determine the fused features as target features.

[0116] The blood vessel risk assessment device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0117] Based on the above embodiment, optionally, the above device further includes: an extraction module, wherein:

[0118] The extraction module is used to extract features from the fused features using a preset feature extraction algorithm, extract nonlinearly related features from the fused features, and determine the extracted features as target features.

[0119] Optionally, the preset feature extraction algorithm includes principal component analysis.

[0120] The blood vessel risk assessment device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0121] Based on the above embodiments, optionally, the preset classification model is a model constructed based on a machine learning algorithm, or a model constructed based on a neural network model including a fully connected layer; the machine learning algorithm includes a random forest method, a support vector machine, a k-nearest neighbor method, a decision tree, a logistic regression model, an XGboost algorithm, a quadratic discriminant analysis method, a stochastic gradient descent method, and an Adaboost algorithm.

[0122] The blood vessel risk assessment device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0123] The specific definition of the vascular risk assessment device can be found in the definition of the vascular risk assessment method above, which will not be repeated here. Each module in the above-mentioned vascular risk assessment device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0124] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0125] Inputting the medical image into a preset segmentation model to obtain a mask image of the artery of interest in the medical image;

[0126] Determine a preset tissue region where the artery of interest is located according to the mask image of the artery of interest;

[0127] Extracting image features of a preset tissue region where an artery of interest is located, and obtaining imaging omics features of the preset tissue region where the artery of interest is located;

[0128] Based on the imaging features and the preset classification model, the risk assessment results of the arteries of interest are obtained.

[0129] The implementation principle and technical effect of the computer device provided in the above embodiment are similar to those of the above method embodiment, and will not be repeated here.

[0130] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0131] Inputting the medical image into a preset segmentation model to obtain a mask image of the artery of interest in the medical image;

[0132] Determine a preset tissue region where the artery of interest is located according to the mask image of the artery of interest;

[0133] Extracting image features of a preset tissue region where an artery of interest is located, and obtaining imaging omics features of the preset tissue region where the artery of interest is located;

[0134] Based on the imaging features and the preset classification model, the risk assessment results of the arteries of interest are obtained.

[0135] The computer-readable storage medium provided in the above embodiment has similar implementation principles and technical effects as those of the above method embodiment, and will not be described in detail here.

[0136] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0137] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0138] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for assessing blood vessel risk, characterized in that: The method comprises: Inputting the medical image into a preset segmentation model to obtain a mask image of an artery of interest in the medical image; Determine a preset tissue region where the artery of interest is located according to the mask image of the artery of interest; wherein the lumen diameter of the artery of interest is enlarged by a preset enlargement coefficient, and the region within the enlarged lumen diameter is determined as the preset tissue region where the artery of interest is located, and the preset tissue region includes the artery of interest; Extracting image features of a preset tissue region where the artery of interest is located, and obtaining imaging omics features of the preset tissue region where the artery of interest is located; According to the radiomics features and the preset classification model, a risk assessment result of the artery of interest is obtained; the risk assessment result of the artery of interest is used to characterize a characteristic assessment result of a plaque in the artery of interest; The method further comprises: Acquire a feature map obtained during the process of segmenting the artery of interest in the medical image by the segmentation model; Obtaining a target feature according to the feature map and the radiomics feature; Inputting the target feature into the classification model to obtain a risk assessment result of the artery of interest; Wherein, obtaining the target feature according to the feature map and the radiomics feature includes: Performing feature conversion on the feature map to obtain converted features; the dimension of the converted features is the same as the dimension of the radiomics features; Performing feature fusion on the converted features and the radiomics features, and determining the fused features as the target features; Wherein, determining the fused feature as the target feature includes: The fused features are extracted using a preset feature extraction algorithm to extract nonlinearly related features from the fused features, and the extracted features are determined as the target features.

2. The method according to claim 1, characterized in that The step of determining the preset tissue region where the artery of interest is located according to the mask image of the artery of interest includes: Obtaining the lumen diameter of the arterial vessel of interest according to the mask image of the arterial vessel of interest; The lumen diameter of the artery of interest is enlarged, and the area within the enlarged lumen diameter is determined as the preset tissue area where the artery of interest is located.

3. The method according to claim 2, characterized in that The step of obtaining the lumen diameter of the arterial vessel of interest according to the mask image of the arterial vessel of interest comprises: Extracting the center line of the arterial vessel of interest from the mask image of the arterial vessel of interest; For each point on the centerline of the artery of interest, determine the cross section where each point is located; The contour of the arterial vessel of interest is extracted based on the cross section, and the lumen diameter of the arterial vessel of interest is obtained according to the contour of the arterial vessel of interest.

4. The method according to claim 1, characterized in that The preset feature extraction algorithm includes principal component analysis.

5. The method according to claim 1, characterized in that The preset classification model is a model constructed based on a machine learning algorithm, or a model constructed based on a neural network model including a fully connected layer; the machine learning algorithm includes a random forest method, a support vector machine, a k-nearest neighbor method, a decision tree, a logistic regression model, an XGboost algorithm, a quadratic discriminant analysis method, a stochastic gradient descent method, and an Adaboost algorithm.

6. A blood vessel risk assessment device, characterized in that: The device comprises: A first acquisition module is used to input the medical image into a preset segmentation model to obtain a mask image of an artery of interest in the medical image; A determination module, configured to determine a preset tissue region where the artery of interest is located according to the mask image of the artery of interest; wherein the lumen diameter of the artery of interest is enlarged by a preset enlargement coefficient, and an area within the enlarged lumen diameter is determined as the preset tissue region where the artery of interest is located, and the preset tissue region includes the artery of interest; An extraction module, used to extract image features of a preset tissue region where the artery of interest is located, and obtain image omics features of the preset tissue region where the artery of interest is located; A second acquisition module is used to obtain a risk assessment result of the artery of interest based on the radiomics feature and a preset classification model; the risk assessment result of the artery of interest is used to characterize a characteristic assessment result of a plaque in the artery of interest; A third acquisition module is used to acquire a feature map obtained in the process of segmenting the artery of interest in the medical image by the segmentation model; A fourth acquisition module, used for obtaining a target feature according to the feature map and the radiomics feature; a fifth acquisition module, configured to input the target feature into the classification model to obtain a risk assessment result of the artery of interest; Wherein, the fourth acquisition module includes: a second acquisition unit and a fusion unit; The second acquisition unit is used to perform feature conversion on the feature map to obtain converted features; the dimension of the converted features is the same as the dimension of the radiomics features; The fusion unit is used to perform feature fusion on the converted feature and the radiomics feature, and determine the fused feature as the target feature; The device also includes: an extraction module; The extraction module is used to perform feature extraction on the fused features using a preset feature extraction algorithm, extract nonlinearly related features from the fused features, and determine the extracted features as the target features.

7. The device according to claim 6, characterized in that The determining module comprises: A first acquisition unit, configured to obtain a lumen diameter of the arterial vessel of interest according to the mask image of the arterial vessel of interest; The determination unit is used to expand the lumen diameter of the artery of interest and determine the area within the expanded lumen diameter as the preset tissue area where the artery of interest is located.

8. The device according to claim 7, characterized in that The first acquisition unit is used to extract the center line of the artery of interest from the mask image of the artery of interest; determine the cross-section of each point on the center line of the artery of interest; extract the contour of the artery of interest based on the cross-section, and obtain the lumen diameter of the artery of interest according to the contour of the artery of interest.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images