Vascular wall state evaluation method, device, equipment and medium

Abstract

The invention discloses a blood vessel wall state evaluation method and device, equipment and a medium, and relates to the technical field of computers, and the method comprises the steps: obtaining a to-be-evaluated blood vessel wall image, carrying out the geometric structure reconstruction of the to-be-evaluated blood vessel wall image through a preset image processing method, and constructing a geometric model of a blood vessel wall; performing mechanical analysis on the geometric model by utilizing a finite element analysis method and based on a preset pressure condition and a preset boundary condition on the blood vessel wall, calculating strain distribution data of blood vessel wall symmetric points at different blood vessel wall positions, and calculating a blood vessel wall strain non-uniformity index based on the strain distribution data; constructing a corresponding joint evaluation index according to the vascular wall strain non-uniformity index, the target mechanical index, the target morphological index and user basic information; and inputting the joint evaluation index into a preset blood vessel wall state evaluation model, and outputting a corresponding blood vessel wall rupture state risk probability through the preset blood vessel wall state evaluation model. And the vascular wall rupture risk is accurately predicted from different angles.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device, equipment and medium for evaluating the state of a blood vessel wall. Background Art

[0002] In the prior art, vascular mechanics analysis mainly focuses on average strain or maximum strain. Although these traditional indicators have certain value in evaluating the strain distribution of vascular walls, they often ignore the asymmetry of strain distribution. Specific defects include: Ignoring local strain differences: Average strain and maximum strain cannot reflect the strain differences at different locations of the vascular wall, especially the strain differences at symmetrical points, and fail to reveal potential structural heterogeneity. Lack of ability to identify diseased areas: Traditional indicators have difficulty in identifying potential vascular wall weaknesses and diseased areas, affecting the accuracy of clinical diagnosis. Insufficient early warning capabilities: In the early stages of diseases such as arteriosclerosis and aneurysms, traditional indicators cannot provide effective warning signals, increasing the possibility of worsening lesions. Strain peak fluctuations caused by individual differences and simulation model settings: affecting early clinical decision-making.

[0003] In summary, how to comprehensively focus on the strain distribution information of the vascular wall to accurately assess the state of the vascular wall is a technical problem to be solved in this field. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide a method, device, equipment and medium for evaluating the state of the vascular wall, which can comprehensively focus on the strain distribution information of the vascular wall to accurately evaluate the state of the vascular wall. The specific scheme is as follows:

[0005] In a first aspect, the present application discloses a method for evaluating a blood vessel wall state, comprising:

[0006] Acquire an image of a blood vessel wall to be evaluated, and use a preset image processing method to reconstruct the geometric structure of the image of the blood vessel wall to be evaluated, so as to construct a geometric model of the blood vessel wall; wherein the geometric model includes the morphological features of the blood vessel wall;

[0007] Performing mechanical analysis on the geometric model using a finite element analysis method based on preset pressure conditions and preset boundary conditions on the blood vessel wall to calculate strain distribution data of symmetric points of the blood vessel wall at different blood vessel wall positions, and calculating a corresponding blood vessel wall strain non-uniformity index based on the strain distribution data;

[0008] Constructing a corresponding joint evaluation index according to the vascular wall strain non-uniformity index, the target mechanical index, the target morphological index and the user basic information corresponding to the vascular wall image to be evaluated; wherein the target mechanical index is a strain-based derived mechanical index;

[0009] The combined evaluation index is input into a preset vascular wall state evaluation model so that the preset vascular wall state evaluation model can output the corresponding vascular wall rupture state risk probability through probability calculation.

[0010] Optionally, the geometric structure reconstruction of the to-be-evaluated vascular wall image by using a preset image processing method to construct a geometric model of the vascular wall includes:

[0011] The image of the blood vessel wall to be evaluated is subjected to image segmentation and three-dimensional modeling processing by using a preset image processing method, so as to perform geometric structure reconstruction of the blood vessel wall and construct a geometric model of the blood vessel wall.

[0012] Optionally, the mechanical analysis of the geometric model using the finite element analysis method based on the preset pressure condition and preset boundary condition of the blood vessel wall to calculate the strain distribution data of the symmetric points of the blood vessel wall at different blood vessel wall positions includes:

[0013] Performing mechanical analysis on the geometric model using a finite element analysis method based on preset pressure conditions and preset boundary conditions on the blood vessel wall, so as to obtain circumferential strain data and radial strain data at different blood vessel wall positions at each time point;

[0014] Determine the blood vessel centerline, and select a number of pairs of symmetric points on the blood vessel wall in the circumferential direction of the blood vessel centerline; the symmetric points on the blood vessel wall include left points and right points on the left and right sides along the direction of the blood vessel centerline;

[0015] The strain distribution data of each pair of symmetrical points on the blood vessel wall are calculated based on the circumferential strain data and the radial strain data.

[0016] Optionally, the calculating a corresponding vascular wall strain non-uniformity index based on the strain distribution data includes:

[0017] The blood vessel wall strain non-uniformity index is calculated according to the first strain distribution data of the left point and the second strain distribution data of the right point of each pair of symmetrical points of the blood vessel wall.

[0018] Optionally, constructing a corresponding joint evaluation index according to the vascular wall strain non-uniformity index, the target mechanical index, the target morphological index and the user basic information corresponding to the vascular wall image to be evaluated includes:

[0019] Substituting the user's blood pressure information and the user's baseline parameter information corresponding to the vascular wall image to be evaluated into the vascular wall strain uniformity coefficient calculation equation to obtain the target vascular wall strain non-uniformity index fused with the basic information;

[0020] Obtaining a maximum vascular wall strain concentration index according to a maximum vascular wall strain non-uniformity index, a minimum vascular wall strain non-uniformity index, and an average vascular wall strain uniformity index of the target vascular wall region, and obtaining a vascular wall strain gradient intensity according to the circumferential strain data and the radial strain data;

[0021] A joint evaluation index is constructed based on the target vascular wall strain heterogeneity index, the maximum vascular wall strain concentration index, the vascular wall strain gradient intensity and the target morphological index after nonlinear mathematical transformation and the corresponding weight coefficients.

[0022] Optionally, the obtaining of the maximum vascular wall strain concentration index according to the maximum vascular wall strain non-uniformity index, the minimum vascular wall strain non-uniformity index and the average vascular wall strain uniformity index of the target vascular wall region, and the obtaining of the vascular wall strain gradient intensity according to the circumferential strain data and the radial strain data, comprises:

[0023] Selecting all the symmetric points of the blood vessel wall in the target blood vessel region as target blood vessel wall symmetric points, so as to determine a maximum blood vessel wall strain non-uniformity index and a minimum blood vessel wall strain non-uniformity index from strain distribution data in the target blood vessel wall symmetric points;

[0024] The average vascular wall strain uniformity index of the target vascular region is determined according to the strain distribution data of each symmetrical point of the target vascular wall, so as to determine the maximum vascular wall strain concentration index according to the maximum strain concentration index equation; wherein the maximum strain concentration index equation is:

[0025] ;

[0026] Among them, MSCI represents the maximum vascular wall strain concentration index, represents the maximum vessel wall strain heterogeneity index, represents the minimum vessel wall strain heterogeneity index, represents the average vessel wall strain uniformity index;

[0027] The vascular wall strain gradient strength is calculated according to the circumferential strain data and the radial strain data by using the strain gradient strength equation; wherein the strain gradient strength equation is:

[0028] ;

[0029] Where SGI represents the vascular wall strain gradient intensity, represents the gradient, represents radial strain data and hoop strain data, Represents the norm.

[0030] Optionally, before inputting the combined evaluation index into a preset vascular wall state evaluation model so as to output the corresponding vascular wall rupture state risk probability through probability calculation by the preset vascular wall state evaluation model, the method further includes:

[0031] A preset vascular wall state assessment model was constructed, which included a regression sub-model for evaluating the association between various evaluation indicators of the vascular wall and the risk of vascular wall rupture, a support vector regression sub-model for processing the nonlinear relationship between various evaluation indicators to output the prediction value of vascular wall rupture risk, and a Bayesian network sub-model for reflecting the condition of vascular wall lesions.

[0032] In a second aspect, the present application discloses a device for evaluating a blood vessel wall state, comprising:

[0033] A structure reconstruction module, used to obtain an image of a blood vessel wall to be evaluated, and to perform geometric structure reconstruction on the image of the blood vessel wall to be evaluated using a preset image processing method, so as to construct a geometric model of the blood vessel wall; wherein the geometric model includes morphological features of the blood vessel wall;

[0034] a strain data acquisition module, configured to perform mechanical analysis on the geometric model using a finite element analysis method based on preset pressure conditions and preset boundary conditions on the blood vessel wall, so as to calculate strain distribution data of symmetric points of the blood vessel wall at different blood vessel wall positions, and calculate a corresponding blood vessel wall strain non-uniformity index based on the strain distribution data;

[0035] A parameter fusion module, used to construct a corresponding joint evaluation index according to the vascular wall strain non-uniformity index, the target mechanical index, the target morphological index and the user basic information corresponding to the vascular wall image to be evaluated; wherein the target mechanical index is a strain-based derived mechanical index;

[0036] The state prediction module is used to input the combined evaluation index into a preset vascular wall state evaluation model so as to output the corresponding vascular wall rupture state risk probability through probability calculation through the preset vascular wall state evaluation model.

[0037] In a third aspect, the present application discloses an electronic device, comprising:

[0038] Memory, used to store computer programs;

[0039] The processor is used to execute the computer program to implement the steps of the aforementioned method for evaluating the state of the blood vessel wall.

[0040] In a fourth aspect, the present application discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the steps of the aforementioned method for evaluating the state of a blood vessel wall are implemented.

[0041] It can be seen that the present application discloses a method for evaluating the state of a blood vessel wall, comprising: obtaining a blood vessel wall image to be evaluated, and reconstructing the geometric structure of the blood vessel wall image to be evaluated by using a preset image processing method to construct a geometric model of the blood vessel wall; wherein the geometric model includes the morphological features of the blood vessel wall; using a finite element analysis method and based on the preset pressure conditions and preset boundary conditions of the blood vessel wall to perform mechanical analysis on the geometric model to calculate the strain distribution data of the symmetric points of the blood vessel wall at different blood vessel wall positions, and calculating the corresponding blood vessel wall strain non-uniformity index based on the strain distribution data; constructing a corresponding joint evaluation index according to the blood vessel wall strain non-uniformity index, the target mechanical index, the target morphological index and the basic information of the user corresponding to the blood vessel wall image to be evaluated; wherein the target mechanical index is a strain-based derivative mechanical index; inputting the joint evaluation index into a preset blood vessel wall state evaluation model, so as to output the corresponding blood vessel wall rupture state risk probability through probability calculation through the preset blood vessel wall state evaluation model. It can be seen that the present application provides a new index reflecting the integrity of the blood vessel structure by quantifying the symmetry of the blood vessel wall strain distribution. Compared with traditional average strain or maximum strain indicators, the vascular wall strain heterogeneity index can better identify potential weaknesses and diseased areas of the vascular wall, providing more valuable support for clinical diagnosis and treatment. Combined with multiple mechanical indicators and morphological indicators, a joint index is constructed. This multi-parameter fusion method can comprehensively evaluate the degree of vascular wall disease and rupture risk from multiple angles, improving the accuracy and reliability of the assessment. Finally, by presetting the vascular wall state assessment model, the risk of vascular wall rupture can be predicted from different angles. Through model fusion and cross-validation, the best performing model can be selected or model fusion can be performed to improve the accuracy and reliability of the prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0043] Figure 1 This is a flow chart of a method for evaluating the state of a blood vessel wall disclosed in the present application;

[0044] Figure 2 A flowchart of a specific joint index construction method disclosed in this application;

[0045] Figure 3 This is a schematic diagram of the structure of a blood vessel wall state assessment device disclosed in this application;

[0046] Figure 4 This is a structural diagram of an electronic device disclosed in this application. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0048] In the prior art, vascular mechanics analysis mainly focuses on average strain or maximum strain. Although these traditional indicators have certain value in evaluating the strain distribution of vascular walls, they often ignore the asymmetry of strain distribution. Specific defects include: Ignoring local strain differences: Average strain and maximum strain cannot reflect the strain differences at different locations of the vascular wall, especially the strain differences at symmetrical points. Lack of ability to identify diseased areas: Traditional indicators have difficulty identifying potential vascular wall weaknesses and diseased areas, affecting the accuracy of clinical diagnosis. Insufficient early warning capabilities: In the early stages of diseases such as arteriosclerosis and aneurysms, traditional indicators cannot provide effective warning signals, increasing the possibility of worsening lesions. Strain peak fluctuations caused by individual differences and simulation model settings: affecting early clinical decision-making.

[0049] To this end, the present invention provides a vascular wall state assessment solution that can comprehensively focus on the strain distribution information of the vascular wall to accurately assess the state of the vascular wall.

[0050] Reference Figure 1 As shown, the embodiment of the present application discloses a method for evaluating the state of a blood vessel wall, comprising:

[0051] Step S11: acquiring a vessel wall image to be evaluated, and reconstructing the geometric structure of the vessel wall image to be evaluated using a preset image processing method to construct a geometric model of the vessel wall; wherein the geometric model includes morphological features of the vessel wall.

[0052] In this embodiment, high-resolution medical imaging technology (such as CT, MRI) is used to obtain the image of the vascular wall to be evaluated; the image of the vascular wall to be evaluated is subjected to image segmentation and three-dimensional modeling processing using a preset image processing method to perform geometric structure reconstruction of the vascular wall to construct a geometric model of the vascular wall. It can be understood that the geometric structure reconstruction and strain distribution data of the vascular wall are performed using image processing technology and finite element simulation technology. Specifically, the image processing technology is used to perform image processing on the image of the vascular wall to be evaluated, wherein the image processing includes operations such as removing noise and enhancing image contrast to improve image quality and facilitate subsequent analysis. Next, the geometric structure reconstruction of the vascular wall is performed using finite element simulation technology. By segmenting and three-dimensionally modeling the image, an accurate geometric model of the vascular wall is constructed. This model will provide a basis for subsequent strain distribution calculations.

[0053] In this embodiment, in the reconstruction of the geometric structure of the vascular wall, the finite element simulation technology is mainly used to simulate the mechanical behavior of the vascular wall under different physiological conditions, so as to reconstruct its geometric structure. Specifically, according to the acquired vascular wall image to be evaluated, an initial geometric model of the vascular wall is established. This initial geometric model usually includes information such as the shape, size, thickness, etc. of the vascular wall. Then the initial geometric model is discretized into a number of small units, each of which has a specific shape and mechanical properties. Common unit types include triangles, quadrilaterals, tetrahedrons, hexahedrons, etc. Determine the connection mode and node position between the units, and construct a finite element mesh. The vascular wall is usually regarded as a nonlinear elastic material, and its mechanical properties can be obtained through experimental measurements or related data. Give each unit corresponding material properties. These material properties will determine the deformation behavior of the vascular wall when it is subjected to force. Further, to simulate the stress of the vascular wall in the body, appropriate loading and boundary conditions need to be applied. The loading conditions may include blood pressure in the blood vessel, blood flow impact force, etc. Blood pressure is usually loaded on the inner wall of the blood vessel as a uniformly distributed pressure. Boundary conditions may include fixed constraints at both ends of the blood vessel, the interaction between the blood vessel and the surrounding tissue, etc. According to the basic principles of the finite element method, the mechanical equilibrium equation is established. For each unit, the stress and strain within the unit are calculated according to its material properties and stress conditions. By assembling the equations of each unit, the mechanical equilibrium equation system of the entire vascular wall is obtained. The equation system is solved using numerical solution methods, such as the Newton-Raphson method, the conjugate gradient method, etc., to obtain the displacement field and stress field of the vascular wall under given loading and boundary conditions. According to the solved displacement field, the geometric shape of the vascular wall is updated. The displacement of each node will cause the shape of the unit to change, thereby updating the geometric structure of the entire vascular wall. Through continuous iterative calculations until the convergence conditions are met, the final vascular wall geometric structure, that is, the final geometric model, is obtained.

[0054] Step S12: Performing mechanical analysis on the geometric model using finite element analysis method based on preset pressure conditions and preset boundary conditions of the blood vessel wall to calculate strain distribution data of symmetrical points of the blood vessel wall at different blood vessel wall positions, and calculating a corresponding blood vessel wall strain non-uniformity index based on the strain distribution data.

[0055] In this embodiment, the geometric model is mechanically analyzed using the finite element analysis method based on the preset pressure conditions and preset boundary conditions of the blood vessel wall to obtain the circumferential strain data and radial strain data of different blood vessel wall positions at each time point; the blood vessel centerline is determined, and several pairs of symmetrical points of the blood vessel wall are selected in the circumferential direction of the blood vessel centerline; the symmetrical points of the blood vessel wall include left points and right points on the left and right sides of the blood vessel centerline; the strain distribution data of each pair of symmetrical points of the blood vessel wall are calculated based on the circumferential strain data and radial strain data. Specifically, the blood vessel wall strain non-uniformity index is calculated based on the first strain distribution data of the left point and the second strain distribution data of the right point of each pair of symmetrical points of the blood vessel wall. It can be understood that the mechanical analysis of the geometric figure can calculate the circumferential strain of different blood vessel wall positions at each time point. and radial strain , wherein the finite element analysis method is used to calculate the strain values ​​of different vascular wall positions according to the pressure and boundary conditions (blood pressure, fixed constraints at both ends of the blood vessel, and the interaction between the blood vessel and the surrounding tissue) on the vascular wall. Furthermore, the vascular centerline is determined, and several symmetrical points are selected in the circumferential direction of the vascular centerline. According to the geometric shape and symmetry of the blood vessel, the vascular wall can be divided into n pairs of vascular wall symmetrical points in the circumferential direction perpendicular to the centerline of the blood flow cavity. Each pair of vascular wall symmetrical points consists of a left point and a right point. For each pair of vascular wall symmetrical points, its strain value is calculated separately, and the strain values ​​of all symmetrical points on the left and right sides are averaged to obtain and Calculate the mean strain value on the left side of the aneurysm area , that is, the first strain distribution data, calculate the average strain value on the right side of the aneurysm area , which is the second strain distribution data. Then the vascular wall strain non-uniformity index SNI (Strain Non-Uniformity Index) is calculated using the strain non-uniformity index calculation formula, as shown below:

[0056] ;

[0057] Wherein, n represents the total number of selected symmetric points on the vascular wall, and i represents the i-th pair of symmetric points on the vascular wall.

[0058] It can be seen that for the numerator, the sum of the absolute values ​​of the difference between the left strain value and the right strain value is calculated for all n pairs of symmetrical points on the vascular wall. This value reflects the sum of the strain differences between all symmetrical points on the entire vascular wall. When the strain distribution is completely symmetrical, the strain values ​​of each pair of symmetrical points are equal, the difference is zero, and the numerator is zero. As the asymmetry of the strain distribution increases, the strain difference between the symmetrical points increases, and the value of the numerator also increases. For the denominator, for all n pairs of symmetrical points on the vascular wall, the sum of the left strain value and the right strain value is calculated. This value reflects the total strain of all symmetrical points on the entire vascular wall. When the strain distribution is completely symmetrical, the strain values ​​of each pair of symmetrical points are equal, the sum is (assuming that the strain value of the symmetrical point is), and the value of the denominator is. As the asymmetry of the strain distribution increases, the strain difference between the symmetrical points causes the change of the sum, but in general the value of the denominator increases with the increase of strain. Finally, from an overall perspective, when the strain distribution is completely symmetrical, the numerator is zero and SNI is equal to 1, indicating that the strain distribution is completely symmetrical. As the asymmetry of the strain distribution increases, the numerator increases and the denominator changes accordingly, but because the numerator is the difference in absolute values ​​and the growth rate is relatively slow, the value of SNI will gradually decrease and approach 0, indicating that the strain distribution is becoming more and more asymmetric.

[0059] It should be noted that SNI can effectively quantify the symmetry of strain distribution in the vascular wall. The closer the value is to 1, the more symmetrical the strain distribution is; the closer the value is to 0, the more asymmetric the strain distribution is. This provides an important indicator for evaluating the structural integrity and potential lesion areas of the vascular wall. By quantifying the symmetry of the strain distribution of the vascular wall, this indicator provides a new indicator that reflects the structural integrity of the blood vessel, helps identify potential weaknesses and lesion areas of the vascular wall, provides support for clinical diagnosis and treatment, and provides early warning in the early stages of diseases such as arteriosclerosis and aneurysms to prevent serious lesions and rupture risks. Traditional vascular mechanics analysis usually focuses on the average strain or maximum strain, but these indicators may ignore the asymmetry of strain distribution, which is important in predicting vascular lesions and rupture risks.

[0060] Step S13: constructing a corresponding joint evaluation index according to the vascular wall strain non-uniformity index, the target mechanical index, the target morphological index and the user basic information corresponding to the vascular wall image to be evaluated; wherein the target mechanical index is a strain-based derived mechanical index.

[0061] In this embodiment, based on the calculated vascular wall strain non-uniformity index SNI, maximum vascular wall strain concentration index (Maximum Strain Concentration Index, MSCI), vascular wall strain gradient intensity (StrainGradientIntensity, SGI) and related user basic information (blood pressure, height, age, medical history, etc.), the above information is unified as vascular wall strain data, and then standardized through nonlinear mathematical transformation, thereby realizing the fusion of multiple parameters and obtaining a joint evaluation index.

[0062] Step S14: inputting the combined evaluation index into a preset vascular wall state assessment model, so as to output the corresponding vascular wall rupture state risk probability through probability calculation by the preset vascular wall state assessment model.

[0063] In this embodiment, the input of the combined evaluation index into the preset vascular wall state evaluation model so that the corresponding vascular wall rupture state risk probability is output through probability calculation by the preset vascular wall state evaluation model also includes: constructing a regression sub-model including the association between various evaluation indicators for evaluating the vascular wall and the risk of vascular wall rupture, a support vector regression sub-model for processing the nonlinear relationship between various evaluation indicators to output the predicted value of vascular wall rupture risk, and a preset vascular wall state evaluation model of the Bayesian network sub-model for reflecting the vascular wall lesion. It can be understood that the regression sub-model is obtained by training based on the Logistics regression model. Specifically, the regression sub-model predicts the risk of vascular wall rupture: by inputting a feature vector including multiple parameters such as SNI, MSCI, and SGI, the probability of vascular wall rupture or non-rupture is output. Evaluate the possibility of vascular wall rupture under specific circumstances so as to take intervention measures in time. The training process is specifically to train and establish the relationship between multiple parameters and rupture risk: using known mechanical indicators and other characteristics, determine the regression coefficient, so as to establish a quantitative relationship between these parameters and the risk of vascular wall rupture. This helps to have a deeper understanding of the influence of different factors on vascular wall rupture. The support vector regression submodel is obtained by training the support vector regression model. Specifically, the support vector regression submodel is used for nonlinear data regression analysis and is suitable for processing the nonlinear relationship between indicators in the evaluation of vascular lesions. The input data is mapped to a high-dimensional space through a kernel function for linear regression, which can more accurately predict the risk of vascular wall rupture, especially in the case of complex data relationships. In the case of limited clinical data, the support vector regression model can more effectively use data for prediction and provide reliable lesion degree and rupture risk assessment results. The Bayesian network submodel is obtained by training the Bayesian network model. It can construct a Bayesian network model based on the selected feature parameters, integrate the information of different parameters, and provide a more comprehensive risk assessment. Considering the dependency between variables, it can more accurately reflect the actual situation of vascular wall lesions. Input aneurysm sample data, use the Bayesian network submodel for reasoning, and output the lesion degree and rupture risk prediction results of each sample. It provides an important basis for clinical diagnosis and treatment decisions, and helps doctors provide effective early warnings in the early stages of diseases such as arteriosclerosis and aneurysms. In this way, the three models together constitute a comprehensive vascular wall rupture risk assessment system. The logistics regression sub-model, support vector regression sub-model and Bayesian network sub-model use the information fused from multiple parameters from different perspectives to predict and evaluate the degree of vascular wall lesions and the risk of rupture.

[0064] The logistics regression sub-model quickly gives a preliminary judgment on the probability of rupture by establishing a linear relationship between multiple parameters and rupture risk. The support vector regression sub-model conducts in-depth analysis of possible nonlinear data relationships to improve the accuracy of risk prediction in complex situations. The Bayesian network sub-model fully considers the dependency between parameters and integrates multi-parameter information for comprehensive risk assessment, providing a more reliable basis for clinical diagnosis and treatment. The three complement and work synergistically to provide more accurate and comprehensive vascular wall rupture risk assessment results so that timely intervention measures can be taken to prevent serious lesions and rupture risks.

[0065] In this embodiment, after the joint evaluation index is input into the preset vascular wall state assessment model, the Logistic sub-regression model has a relatively simple mathematical form and clear interpretability, and can quickly give the characteristics of the probability estimation of the risk of vascular wall rupture. When the relationship between the variables in the joint evaluation index is approximately linear, the Logistic regression sub-model can better capture this relationship and provide relatively accurate prediction results. For the complex nonlinear relationship that may exist in the joint evaluation index, the support vector regression sub-model can map the data to a high-dimensional space for linear regression through a kernel function, effectively process nonlinear data, and improve the accuracy of the prediction. Therefore, in the case of limited clinical data, the support vector regression sub-model can better utilize limited data for learning and prediction, and has a strong generalization ability. The Bayesian network sub-model is based on probability theory and can perform probabilistic reasoning on the joint evaluation index. It can not only give the prediction results of the degree of lesions and the risk of rupture, but also provide uncertainty measurement, providing doctors with more comprehensive information. By constructing a probabilistic graph model, the Bayesian network can integrate the information of different parameters in the joint evaluation index, consider the dependency between variables, and more accurately reflect the actual situation of vascular wall lesions.

[0066] In this embodiment, before the model outputs the corresponding vascular wall lesion prediction results, the corresponding corresponding relationship is constructed in advance according to the actual results of different vascular wall lesions of the user and the corresponding different SNI numerical results calculated in combination with other parameter information. Specifically, the healthy blood vessel result: the strain distribution of healthy blood vessels is relatively uniform. Although completely symmetrical strain distribution is not common due to geometric shape and material properties, its strain symmetry index SNI still reflects the normal structure and function of the vascular wall. Mild lesion results: In the case of early arteriosclerosis or mild lesions, local strain asymmetry may occur in the vascular wall. At this time, SNI decreases slightly, but the blood vessel as a whole is still relatively stable, indicating slight hardening or decreased elasticity. Moderate lesion results: In the early stage of aneurysm or moderate arteriosclerosis, the strain asymmetry of the vascular wall increases. SNI decreases significantly, indicating that there are structural weaknesses in the lesion area, which requires further monitoring or intervention. Severe lesion results: In the case of aneurysm expansion or arterial dissection, the strain of the vascular wall is severely asymmetric. SNI decreases significantly, the strain in the lesion area is concentrated, the structure is severely damaged, and the risk of rupture is high, and treatment is required as soon as possible. Extreme lesion results: In cases of severe arterial dissection or impending rupture of aneurysm, SNI is close to the lowest value, reflecting a highly asymmetric strain distribution, indicating that the vessel wall is about to rupture and requires immediate intervention.

[0067] It can be seen that the present application discloses a method for evaluating the state of a blood vessel wall, comprising: obtaining a blood vessel wall image to be evaluated, and reconstructing the geometric structure of the blood vessel wall image to be evaluated by using a preset image processing method to construct a geometric model of the blood vessel wall; wherein the geometric model includes the morphological features of the blood vessel wall; using a finite element analysis method and based on the preset pressure conditions and preset boundary conditions of the blood vessel wall to perform mechanical analysis on the geometric model to calculate the strain distribution data of the symmetric points of the blood vessel wall at different blood vessel wall positions, and calculating the corresponding blood vessel wall strain non-uniformity index based on the strain distribution data; constructing a corresponding joint evaluation index according to the blood vessel wall strain non-uniformity index, the target mechanical index, the target morphological index and the basic information of the user corresponding to the blood vessel wall image to be evaluated; wherein the target mechanical index is a strain-based derivative mechanical index; inputting the joint evaluation index into a preset blood vessel wall state evaluation model, so as to output the corresponding blood vessel wall rupture state risk probability through probability calculation through the preset blood vessel wall state evaluation model. It can be seen that the present application provides a new index reflecting the integrity of the blood vessel structure by quantifying the symmetry of the blood vessel wall strain distribution. Compared with traditional average strain or maximum strain indicators, the vascular wall strain heterogeneity index can better identify potential weaknesses and diseased areas of the vascular wall, providing more valuable support for clinical diagnosis and treatment. Combined with multiple mechanical indicators and morphological indicators, a joint index is constructed. This multi-parameter fusion method can comprehensively evaluate the degree of vascular wall disease and rupture risk from multiple angles, improving the accuracy and reliability of the assessment. Finally, by presetting the vascular wall state assessment model, the risk of vascular wall rupture can be predicted from different angles. Through model fusion and cross-validation, the best performing model can be selected or model fusion can be performed to improve the accuracy and reliability of the prediction.

[0068] The embodiment of the present invention discloses a specific method for constructing a joint evaluation index. Compared with the previous embodiment, this embodiment further illustrates and optimizes the technical solution. Specifically:

[0069] Step S21: Substituting the user blood pressure information and the user baseline parameter information corresponding to the vascular wall image to be evaluated into the vascular wall strain uniformity coefficient calculation equation to obtain the target vascular wall strain non-uniformity index fused with the basic information.

[0070] In this embodiment, the user's blood pressure information and the user's baseline parameter information are combined with the vascular wall strain uniformity coefficient calculation equation. Therefore, the calculation formula of SNI is updated as follows:

[0071] ;

[0072] in, Indicates the user's blood pressure information. represents the weight coefficient of the corresponding i-th pair of symmetric points, The weight function representing the user's baseline parameter information, Represents a set of baseline parameters for the patient, including height, weight, age, and medical history.

[0073] represents the strain value of the right point in the i-th pair of symmetric points of the blood vessel wall, represents the strain value of the left point in the i-th pair of symmetric points of the vascular wall, Represents the average strain value of the area near the symmetric point of the i-th pair of blood vessel walls.

[0074] In this way, since blood pressure is one of the important factors affecting the strain of the vascular wall, different blood pressure levels will produce different degrees of pressure on the vascular wall, thereby affecting the strain distribution. The new formula incorporates blood pressure into the calculation, which can more accurately reflect the strain of the vascular wall under actual physiological conditions. For example, higher blood pressure may cause the vascular wall to bear greater stress, thereby increasing the strain non-uniformity. By considering the blood pressure factor, the risk of vascular wall lesions can be more comprehensively assessed. Baseline parameters such as height, weight, age and previous medical history are also closely related to the health of the vascular wall. Different individuals have different physiological characteristics and disease risk factors, which affect the mechanical properties of the vascular wall and the development of lesions. The weight function in the new formula takes these baseline parameters into account, and can perform personalized assessments based on individual differences in patients. For example, patients who are older, have a history of disease or are obese may be more likely to develop vascular lesions. By considering these factors, the risk of vascular wall rupture in these patients can be more accurately predicted.

[0075] Step S22: obtaining a maximum vascular wall strain concentration index according to the maximum vascular wall strain non-uniformity index, the minimum vascular wall strain non-uniformity index and the average vascular wall strain uniformity index of the target vascular wall area, and obtaining the vascular wall strain gradient intensity according to the circumferential strain data and the radial strain data.

[0076] In this embodiment, all the symmetric points of the blood vessel wall in the target blood vessel region are selected as the target blood vessel wall symmetric points, so as to determine the maximum blood vessel wall strain non-uniformity index and the minimum blood vessel wall strain non-uniformity index from the strain distribution data in the target blood vessel wall symmetric points; the average blood vessel wall strain uniformity index of the target blood vessel region is determined according to the strain distribution data of each of the target blood vessel wall symmetric points, so as to determine the maximum blood vessel wall strain concentration index according to the maximum strain concentration index equation; wherein the maximum strain concentration index equation is:

[0077] ;

[0078] Among them, MSCI represents the maximum vascular wall strain concentration index, represents the maximum vessel wall strain heterogeneity index, represents the minimum vessel wall strain heterogeneity index, represents the average vessel wall strain uniformity index;

[0079] The vascular wall strain gradient strength is calculated according to the circumferential strain data and the radial strain data by using the strain gradient strength equation; wherein the strain gradient strength equation is:

[0080] ;

[0081] Where SGI represents the vascular wall strain gradient intensity, represents the gradient, represents radial strain data and hoop strain data, represents the norm. It is understandable that a larger gradient may indicate local stress concentration. Strain gradient intensity (SGI) reflects the rate of change of strain in space. High strain gradient areas may indicate significant stress concentration in the vessel wall, which is prone to structural damage. By calculating the strain gradient intensity, the change of strain in the vessel wall is quantified. It provides more information about the health of the vessel wall structure and helps predict and prevent the occurrence of lesions.

[0082] Step S23: constructing a joint evaluation index according to the target vascular wall strain non-uniformity index, the maximum vascular wall strain concentration index, the vascular wall strain gradient intensity and the target morphological index after nonlinear mathematical transformation and the corresponding weight coefficients.

[0083] In this embodiment, nonlinear mathematical transformation processing of vascular wall strain data is implemented by exponential transformation or logarithmic transformation. Nonlinear mathematical transformation of each vascular wall strain data can enhance its sensitivity to small changes. In the early stage of vascular lesions, the changes in vascular wall strain data may be very small, and linear transformation may not be able to effectively capture these changes. Through nonlinear transformations such as exponential transformation or logarithmic transformation, the difference in strain parameters can be amplified, making early lesions easier to detect. Moreover, different strain parameters may have different data distribution characteristics. Nonlinear mathematical transformation can convert these parameters into a form that is more suitable for analysis, making them more consistent with the assumptions of a specific model or easier to perform subsequent data processing. For example, some strain parameters may have a long-tail distribution, and through logarithmic transformation, their distribution can be closer to a normal distribution, which is convenient for statistical analysis. Moreover, nonlinear transformation can improve the performance of the model. In some machine learning algorithms, the distribution of input data has a great influence on the training and prediction effects of the model. By performing nonlinear transformation on the parameters, the data can be made more suitable for model learning, improving the accuracy and generalization ability of the model.

[0084] In this embodiment, the vascular wall strain data after nonlinear mathematical transformation is subjected to data dimensionality reduction processing, that is, other features of the high-dimensional vascular wall strain data are converted into a low-dimensional space, thereby reducing the amount of calculation and storage requirements, improving the calculation efficiency, and removing noise and redundant information, retaining relevant information valuable for the prediction of vascular lesions, and finally subjecting the retained information to multi-parameter fusion to construct a joint evaluation index.

[0085] It can be seen that performing nonlinear mathematical transformation of parameters, data dimension reduction and multi-parameter fusion steps can enhance the sensitivity of strain parameters, reduce computational complexity, remove noise and redundant information, comprehensively consider the impact of multiple factors on vascular lesions, improve the generalization ability of the model and support personalized medicine, thereby providing a more accurate, reliable and practical method for the assessment of the risk of vascular wall rupture.

[0086] Reference Figure 3 As shown, the present invention provides a device for evaluating the state of a blood vessel wall, comprising:

[0087] The structure reconstruction module 11 is used to obtain the image of the blood vessel wall to be evaluated, and use a preset image processing method to perform geometric structure reconstruction on the image of the blood vessel wall to be evaluated, so as to construct a geometric model of the blood vessel wall; wherein the geometric model includes the morphological features of the blood vessel wall;

[0088] a strain data acquisition module 12, configured to perform mechanical analysis on the geometric model using a finite element analysis method based on preset pressure conditions and preset boundary conditions on the blood vessel wall, so as to calculate strain distribution data of symmetric points of the blood vessel wall at different blood vessel wall positions, and calculate a corresponding blood vessel wall strain non-uniformity index based on the strain distribution data;

[0089] The parameter fusion module 13 is used to construct a corresponding joint evaluation index according to the vascular wall strain non-uniformity index, the target mechanical index, the target morphological index and the user basic information corresponding to the vascular wall image to be evaluated; wherein the target mechanical index is a strain-based derived mechanical index;

[0090] The state prediction module 14 is used to input the combined evaluation index into a preset blood vessel wall state evaluation model, so as to output the corresponding blood vessel wall rupture state risk probability through probability calculation through the preset blood vessel wall state evaluation model.

[0091] It can be seen that the present application discloses obtaining a vessel wall image to be evaluated, and reconstructing the geometric structure of the vessel wall image to be evaluated by using a preset image processing method to construct a geometric model of the vessel wall; wherein the geometric model includes the morphological features of the vessel wall; using the finite element analysis method and based on the preset pressure conditions and preset boundary conditions of the vessel wall to perform mechanical analysis on the geometric model to calculate the strain distribution data of the symmetric points of the vessel wall at different positions of the vessel wall, and calculating the corresponding vessel wall strain non-uniformity index based on the strain distribution data; constructing a corresponding joint evaluation index according to the vessel wall strain non-uniformity index, the target mechanical index, the target morphological index and the basic information of the user corresponding to the vessel wall image to be evaluated; wherein the target mechanical index is a strain-based derivative mechanical index; inputting the joint evaluation index into a preset vessel wall state evaluation model, so as to output the corresponding vessel wall rupture state risk probability through probability calculation through the preset vessel wall state evaluation model. It can be seen that the present application provides a new index reflecting the integrity of the vessel structure by quantifying the symmetry of the vessel wall strain distribution. Compared with traditional average strain or maximum strain indicators, the vascular wall strain heterogeneity index can better identify potential weaknesses and diseased areas of the vascular wall, providing more valuable support for clinical diagnosis and treatment. Combined with multiple mechanical indicators and morphological indicators, a joint index is constructed. This multi-parameter fusion method can comprehensively evaluate the degree of vascular wall disease and rupture risk from multiple angles, improving the accuracy and reliability of the assessment. Finally, by presetting the vascular wall state assessment model, the risk of vascular wall rupture can be predicted from different angles. Through model fusion and cross-validation, the best performing model can be selected or model fusion can be performed to improve the accuracy and reliability of the prediction.

[0092] Furthermore, the present application also discloses an electronic device. Figure 4 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content in the diagram cannot be regarded as any limitation on the scope of use of the present application.

[0093] Figure 4 The structure diagram of an electronic device 20 provided in an embodiment of the present application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 is used to store a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the vascular wall state assessment method disclosed in any of the above embodiments. In addition, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0094] In this embodiment, the power supply 23 is used to provide working voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and the external device, and the communication protocol it follows is any communication protocol that can be applied to the technical solution of the present application, and is not specifically limited here; the input and output interface 25 is used to obtain external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs and is not specifically limited here.

[0095] Among them, the processor 21 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 21 can be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 21 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 21 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 21 may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.

[0096] In addition, the memory 22, as a carrier for storing resources, can be a read-only memory, a random access memory, a disk or an optical disk, etc. The resources stored thereon can include an operating system 221, a computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0097] The operating system 221 is used to manage and control the hardware devices and computer programs 222 on the electronic device 20, so as to realize the operation and processing of the massive data 223 in the memory 22 by the processor 21, and it can be Windows Server, Netware, Unix, Linux, etc. In addition to including computer programs that can be used to complete the vascular wall state assessment method performed by the electronic device 20 disclosed in any of the aforementioned embodiments, the computer program 222 can also further include computer programs that can be used to complete other specific tasks. In addition to data transmitted from an external device received by the electronic device, the data 223 can also include data collected by its own input and output interface 25.

[0098] Furthermore, the present application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the aforementioned vascular wall state assessment method is implemented. The specific steps of the method can be referred to the corresponding contents disclosed in the aforementioned embodiments, and will not be repeated here.

[0099] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0100] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application. The steps of the method or algorithm described in conjunction with the embodiments disclosed herein can be implemented directly with hardware, a software module executed by a processor, or a combination of the two. The software module can be placed in a random access memory RAM (Random Access Memory), memory, read-only memory ROM (Read Only Memory), electrically programmable EPROM (Electrically Programmable Read Only Memory), electrically erasable programmable EEPROM (ElectricErasable Programmable Read Only Memory), register, hard disk, removable disk, CD-ROM (CompactDisc-Read Only Memory), or any other form of storage medium known in the technical field.

[0101] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0102] The scheme provided by the present invention is introduced in detail above. Specific examples are used in this article to illustrate the principle and implementation mode of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation mode and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for evaluating a blood vessel wall state, characterized in that: include: Acquire an image of a blood vessel wall to be evaluated, and use a preset image processing method to reconstruct the geometric structure of the image of the blood vessel wall to be evaluated, so as to construct a geometric model of the blood vessel wall; wherein the geometric model includes the morphological features of the blood vessel wall; Performing mechanical analysis on the geometric model using a finite element analysis method based on preset pressure conditions and preset boundary conditions on the blood vessel wall to calculate strain distribution data of symmetric points of the blood vessel wall at different blood vessel wall positions, and calculating a corresponding blood vessel wall strain non-uniformity index based on the strain distribution data; Constructing a corresponding joint evaluation index according to the vascular wall strain non-uniformity index, the target mechanical index, the target morphological index and the user basic information corresponding to the vascular wall image to be evaluated; wherein the target mechanical index is a strain-based derived mechanical index; The combined evaluation index is input into a preset vascular wall state evaluation model so that the preset vascular wall state evaluation model can output the corresponding vascular wall rupture state risk probability through probability calculation.

2. The method for evaluating the state of a blood vessel wall according to claim 1, characterized in that: The method of reconstructing the geometric structure of the image of the blood vessel wall to be evaluated by using a preset image processing method to construct a geometric model of the blood vessel wall includes: The image of the blood vessel wall to be evaluated is subjected to image segmentation and three-dimensional modeling processing by using a preset image processing method, so as to perform geometric structure reconstruction of the blood vessel wall and construct a geometric model of the blood vessel wall.

3. The method for evaluating the state of a blood vessel wall according to claim 1, characterized in that: The method of using the finite element analysis method to perform mechanical analysis on the geometric model based on the preset pressure condition and the preset boundary condition of the blood vessel wall to calculate the strain distribution data of the symmetric points of the blood vessel wall at different blood vessel wall positions includes: Performing mechanical analysis on the geometric model using a finite element analysis method based on preset pressure conditions and preset boundary conditions on the blood vessel wall, so as to obtain circumferential strain data and radial strain data at different blood vessel wall positions at each time point; Determine the blood vessel centerline, and select a number of pairs of symmetric points on the blood vessel wall in the circumferential direction of the blood vessel centerline; the symmetric points on the blood vessel wall include left points and right points on the left and right sides along the direction of the blood vessel centerline; The strain distribution data of each pair of symmetrical points on the blood vessel wall are calculated based on the circumferential strain data and the radial strain data.

4. The method for evaluating the state of a blood vessel wall according to claim 3, characterized in that: The calculating the corresponding blood vessel wall strain non-uniformity index based on the strain distribution data comprises: The blood vessel wall strain non-uniformity index is calculated according to the first strain distribution data of the left point and the second strain distribution data of the right point of each pair of symmetrical points of the blood vessel wall.

5. The method for evaluating the state of a blood vessel wall according to claim 4, characterized in that: The corresponding joint evaluation index is constructed according to the vascular wall strain non-uniformity index, the target mechanical index, the target morphological index and the user basic information corresponding to the vascular wall image to be evaluated, including: Substituting the user's blood pressure information and the user's baseline parameter information corresponding to the vascular wall image to be evaluated into the vascular wall strain uniformity coefficient calculation equation to obtain the target vascular wall strain non-uniformity index fused with the basic information; Obtaining a maximum vascular wall strain concentration index according to a maximum vascular wall strain non-uniformity index, a minimum vascular wall strain non-uniformity index, and an average vascular wall strain uniformity index of the target vascular wall region, and obtaining a vascular wall strain gradient intensity according to the circumferential strain data and the radial strain data; A joint evaluation index is constructed based on the target vascular wall strain heterogeneity index, the maximum vascular wall strain concentration index, the vascular wall strain gradient intensity and the target morphological index after nonlinear mathematical transformation and the corresponding weight coefficients.

6. The method for evaluating the state of a blood vessel wall according to claim 5, characterized in that: The step of obtaining a maximum vascular wall strain concentration index according to a maximum vascular wall strain non-uniformity index, a minimum vascular wall strain non-uniformity index, and an average vascular wall strain uniformity index of the target vascular wall region, and obtaining a vascular wall strain gradient intensity according to the circumferential strain data and the radial strain data, comprises: Selecting all the symmetric points of the blood vessel wall in the target blood vessel region as target blood vessel wall symmetric points, so as to determine a maximum blood vessel wall strain non-uniformity index and a minimum blood vessel wall strain non-uniformity index from strain distribution data in the target blood vessel wall symmetric points; The average vascular wall strain uniformity index of the target vascular region is determined according to the strain distribution data of each symmetrical point of the target vascular wall, so as to determine the maximum vascular wall strain concentration index according to the maximum strain concentration index equation; wherein the maximum strain concentration index equation is: ； Among them, MSCI represents the maximum vascular wall strain concentration index, represents the maximum vessel wall strain heterogeneity index, represents the minimum vessel wall strain heterogeneity index, represents the average vessel wall strain uniformity index; The vascular wall strain gradient strength is calculated according to the circumferential strain data and the radial strain data by using the strain gradient strength equation; wherein the strain gradient strength equation is: ； Where SGI represents the vascular wall strain gradient intensity, represents the gradient, represents radial strain data and hoop strain data, Represents the norm.

7. The method for evaluating the state of a blood vessel wall according to any one of claims 1 to 6, characterized in that: Before inputting the combined evaluation index into a preset vascular wall state evaluation model so as to output the corresponding vascular wall rupture state risk probability through probability calculation by the preset vascular wall state evaluation model, the method further includes: A preset vascular wall state assessment model was constructed, which included a regression sub-model for evaluating the association between various evaluation indicators of the vascular wall and the risk of vascular wall rupture, a support vector regression sub-model for processing the nonlinear relationship between various evaluation indicators to output the prediction value of vascular wall rupture risk, and a Bayesian network sub-model for reflecting the condition of vascular wall lesions.

8. A device for evaluating the state of a blood vessel wall, characterized in that: include: A structure reconstruction module, used to obtain an image of a blood vessel wall to be evaluated, and to perform geometric structure reconstruction on the image of the blood vessel wall to be evaluated using a preset image processing method, so as to construct a geometric model of the blood vessel wall; wherein the geometric model includes morphological features of the blood vessel wall; a strain data acquisition module, configured to perform mechanical analysis on the geometric model using a finite element analysis method based on preset pressure conditions and preset boundary conditions on the blood vessel wall, so as to calculate strain distribution data of symmetric points of the blood vessel wall at different blood vessel wall positions, and calculate a corresponding blood vessel wall strain non-uniformity index based on the strain distribution data; A parameter fusion module, used to construct a corresponding joint evaluation index according to the vascular wall strain non-uniformity index, the target mechanical index, the target morphological index and the user basic information corresponding to the vascular wall image to be evaluated; wherein the target mechanical index is a strain-based derived mechanical index; The state prediction module is used to input the combined evaluation index into a preset vascular wall state evaluation model so as to output the corresponding vascular wall rupture state risk probability through probability calculation through the preset vascular wall state evaluation model.

9. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor is used to execute the computer program to implement the steps of the blood vessel wall state assessment method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: Used to store a computer program; wherein, when the computer program is executed by a processor, the steps of the method for evaluating the state of a blood vessel wall as claimed in any one of claims 1 to 7 are implemented.

Images