A coronary artery vulnerable plaque detection and analysis system

Abstract

The present invention relates to the field of cardiovascular medicine and discloses a coronary artery vulnerable plaque detection and analysis system, comprising: a data collection module for collecting multimodal imaging data of the coronary arteries and identifying the coronary artery vascular structure and type; a plaque region identification module for locating the branch openings of the coronary arteries and analyzing multiple types of coronary plaques; a plaque classification detection module for extracting the structural features of multiple types of plaques, identifying vulnerable and stable plaques in the coronary arteries, and establishing a coronary artery plaque classification detection system; a vulnerable plaque assessment module for constructing a coronary artery vulnerable plaque scoring mechanism based on the plaque vulnerability index, plaque composition, and radial wall strain value; and a result feedback module for combining the vulnerable plaque scoring mechanism with the plaque classification detection system to generate coronary artery vulnerable plaque feedback results. The present invention can improve the accuracy of coronary artery vulnerable plaque detection.

Description

Technical Field

[0001] The present invention relates to a coronary artery vulnerable plaque detection and analysis system, belonging to the field of cardiovascular medicine. Background Art

[0002] Vulnerable coronary plaque refers to an atherosclerotic plaque formed on the wall of the coronary artery, which has a high potential risk of rupture and is extremely likely to cause acute adverse cardiovascular events. Once the plaque ruptures, the procoagulant substances inside it are exposed to the blood, which will quickly lead to acute blockage of the coronary artery, thereby triggering acute coronary syndromes such as angina pectoris and myocardial infarction. According to statistics, cardiovascular disease is one of the leading causes of death worldwide, and acute coronary syndrome caused by rupture of vulnerable coronary plaque is the most serious and common type of cardiovascular disease. In order to gain a deeper understanding of the formation, development and rupture mechanism of acute coronary syndrome, it is necessary to detect vulnerable coronary plaques.

[0003] Traditional coronary artery vulnerable plaque detection mainly relies on imaging examination techniques, such as angiography, to provide morphological data of plaques. Although this method can clearly display the lumen morphology of the coronary artery, it cannot directly detect vulnerable plaque characteristics, such as inflammatory status, which affects the accurate judgment of vulnerable coronary artery plaques.

[0004] Therefore, there is an urgent need for a solution that can improve the accuracy of coronary artery vulnerable plaque detection. Summary of the Invention

[0005] The present invention provides a coronary artery vulnerable plaque detection and analysis system, the main purpose of which is to improve the accuracy of coronary artery vulnerable plaque detection.

[0006] To achieve the above-mentioned objectives, the present invention provides a coronary artery vulnerable plaque detection and analysis system, comprising: a data collection module, a plaque area recognition module, a plaque classification detection module, a vulnerable plaque assessment module, and a result feedback module;

[0007] The data collection module is configured to collect multimodal imaging data of the coronary arteries, identify the vascular structure of the coronary arteries based on the multimodal imaging data, determine the vascular type of the coronary arteries based on the vascular structure, and acquire hemodynamic parameters of the coronary arteries based on the vascular structure and the vascular type;

[0008] The plaque region identification module is configured to locate the branch opening positions of the coronary artery based on the blood vessel type, identify the lesion site of the coronary artery based on the branch opening positions, and separate out multiple types of plaques in the coronary artery based on the lesion site;

[0009] The plaque classification and detection module is configured to extract structural features of the multiple types of plaques, analyze the plaque stability of the coronary artery based on the structural features, identify vulnerable plaques and stable plaques in the coronary artery based on the plaque stability, and establish a plaque classification and detection system for the coronary artery based on the vulnerable plaques and the stable plaques;

[0010] The vulnerable plaque assessment module is configured to extract the plaque composition of the coronary artery from the structural features, measure the radial wall strain value of the coronary artery, calculate the plaque vulnerability index of the coronary artery based on the plaque composition, the radial wall strain value, and the hemodynamic parameter, and construct a vulnerable plaque scoring mechanism for the coronary artery based on the plaque vulnerability index, the plaque composition, and the radial wall strain value;

[0011] The result feedback module is used to generate vulnerable plaque feedback results of the coronary artery by combining the vulnerable plaque scoring mechanism and the plaque classification detection system.

[0012] Optionally, determining the vascular type of the coronary artery according to the vascular structure includes:

[0013] collecting image data of the coronary artery according to the vascular structure;

[0014] extracting a vascular centerline and vascular branches of the coronary artery from the image data;

[0015] identifying a vascular layout of the coronary artery based on the vascular centerline and the vascular branches;

[0016] Calculating the branching angles of the blood vessel branches and the curvature of the blood vessel centerline according to the blood vessel layout;

[0017] Based on the image data, identifying the vascular lumen and vascular outer wall of the coronary artery, and measuring the corresponding inner lumen diameter and outer wall thickness of the vascular lumen and the vascular outer wall;

[0018] extracting multi-dimensional features of the blood vessel structure by combining the branch angle, the degree of curvature, the inner cavity diameter, and the outer wall thickness;

[0019] The vessel type of the coronary artery is determined according to the multidimensional features.

[0020] Optionally, locating the branch opening position of the coronary artery based on the blood vessel type includes:

[0021] Based on the blood vessel type, identifying a branching pattern of the coronary artery, and extracting blood vessel branch points of the coronary artery in the branching pattern;

[0022] identifying the blood supply organ of the coronary artery and extracting organ decomposition features of the blood supply organ;

[0023] constructing a reference coordinate system of the coronary artery based on the organ decomposition characteristics;

[0024] Identifying variant vascular features of the coronary artery according to the vascular type;

[0025] generating a spatial mapping network of the blood vessel types and the blood vessel branch points based on the variant blood vessel characteristics and the branching pattern;

[0026] The branch opening position of the coronary artery is located by combining the spatial mapping network and the reference coordinate system.

[0027] Optionally, the extracting multiple types of plaques in the coronary artery based on the lesion site includes:

[0028] collecting a multimodal vascular image of the coronary artery based on the lesion location;

[0029] identifying the decomposed structure of the coronary artery and extracting the pathological features of the lesion site based on the multimodal vascular image;

[0030] extracting the deposit region of the coronary artery from the multimodal vascular image based on the decomposition structure and the pathological features;

[0031] Identify the grayscale distribution and patch geometry of the sediment area;

[0032] Separating the coronary artery plaque deposits according to the grayscale distribution and the plaque geometry;

[0033] Extracting depth texture features of the patch sediment in the sediment area;

[0034] Combining the grayscale distribution, the geometric shape of the plaque and the depth texture characteristics, multiple categories of plaques in the coronary artery are separated.

[0035] Optionally, analyzing the stability of the coronary artery plaque according to the structural characteristics includes:

[0036] Analyzing the structural complexity of the multi-category plaques according to the structural characteristics, and identifying characteristic elements corresponding to the multi-category plaques;

[0037] performing normalization processing on the characteristic elements to obtain normalized characteristic elements;

[0038] Calculating the probability distribution ratio of the normalized feature elements;

[0039] Based on the probability distribution ratio, calculating the characteristic entropy value corresponding to the normalized characteristic element;

[0040] The stability of the coronary artery plaque is analyzed according to the structural complexity and the characteristic entropy value.

[0041] Optionally, the identifying vulnerable plaques and stable plaques in the coronary arteries based on the plaque stability comprises:

[0042] collecting medical imaging data of the coronary artery, and extracting the vessel wall and plaque deposition area of ​​the coronary artery from the medical imaging data;

[0043] extracting morphological features, compositional features, and hemodynamic features of the plaque deposition area, and extracting geometric features of the blood vessel wall;

[0044] Based on the morphological characteristics, the component characteristics, the hemodynamic characteristics, and the geometric characteristics, a stability score is performed on different plaques in the plaque deposition area to obtain a plaque scoring result;

[0045] According to the plaque stability, setting the stability thresholds of the different plaques;

[0046] Based on the stability threshold and the plaque scoring result, vulnerable plaques and stable plaques in the coronary artery are identified.

[0047] Optionally, the coronary artery plaque classification and detection system is set up according to the vulnerable plaque and the stable plaque, including:

[0048] extracting classified plaque features of the coronary artery according to the vulnerable plaque and the stable plaque;

[0049] detecting blood markers of the coronary arteries, and identifying the levels of inflammatory markers of the coronary arteries based on the blood markers;

[0050] calculating the correlation coefficient between the inflammatory marker level and the classified plaque characteristics;

[0051] analyzing the correlation pattern between the inflammatory marker level and the classified plaque characteristics according to the correlation coefficient;

[0052] Based on the association pattern, a coronary artery plaque classification and detection system is set up.

[0053] Optionally, the calculating the coronary artery plaque vulnerability index by combining the plaque composition, the radial wall strain value, and the hemodynamic parameter includes:

[0054] extracting the lipid core ratio and fibrous cap thickness of the coronary artery based on the plaque composition;

[0055] Analyzing the shear stress, blood flow velocity, and blood flow direction of the coronary artery based on the hemodynamic parameters;

[0056] identifying a vascular oscillation shear index of the coronary artery based on the shear stress, the blood flow velocity, and the blood flow direction;

[0057] The coronary artery plaque vulnerability index is calculated according to the lipid core ratio, the fibrous cap thickness, the shear stress, and the vascular oscillation shear index using the following formula:

[0058] I=w1·L+w2·(1-C)+w3·S+w4·(1-B)+w5·O

[0059] Among them, I represents the coronary artery plaque vulnerability index, w1·L represents the contribution value of the lipid core ratio in the plaque component, w2·(1-C) represents the contribution value corresponding to the fibrous cap thickness in the plaque component, w3·S represents the contribution value corresponding to the radial wall strain value, w4·(1-B) represents the contribution value corresponding to the shear stress in the hemodynamic parameters, and w5·O represents the contribution value corresponding to the vascular oscillation shear index.

[0060] Optionally, constructing a vulnerable plaque scoring mechanism for the coronary artery based on the plaque vulnerability index, the plaque composition, and the radial wall strain value includes:

[0061] defining a plaque vulnerability evaluation index of the coronary artery according to the plaque vulnerability index, the plaque composition, and the radial wall strain value;

[0062] Calculating the coronary artery plaque stability score based on the plaque vulnerability evaluation index;

[0063] Identifying a numerical range of the plaque stability score, and defining a grading threshold of the plaque stability score based on the numerical range;

[0064] Setting the plaque damage grade of the coronary artery based on the grade threshold;

[0065] extracting plaque features corresponding to the plaque vulnerability evaluation index, and identifying the plaque category corresponding to the plaque features;

[0066] Creating a result interpretation of the plaque score according to the plaque characteristics, the plaque category, and the plaque damage grade;

[0067] Based on the plaque stability score and the result interpretation, a plaque vulnerability score mechanism for the coronary artery is generated.

[0068] Optionally, the generating of the vulnerable plaque feedback result of the coronary artery by combining the vulnerable plaque scoring mechanism and the plaque classification detection system includes:

[0069] Based on the plaque classification detection system, identifying the classified plaques of the coronary artery;

[0070] collecting sequential images of the classified plaques, and extracting historical change trends of the classified plaques from the sequential images;

[0071] generating a disease progression report of the classified plaque according to the historical change trend;

[0072] outputting a plaque score of the classified plaque based on the vulnerable plaque scoring mechanism;

[0073] generating personalized feedback on the classified plaques based on the plaque score and the disease progression report;

[0074] Based on the disease progression report and the personalized feedback, a feedback result of the vulnerable plaque of the coronary artery is output.

[0075] Compared with the problems described in the background technology, the embodiments of the present invention collect multimodal imaging data of the coronary arteries and identify the vascular structure and vascular type of the coronary arteries, thereby avoiding misdiagnosis or missed diagnosis that may occur in single-modal imaging, helping to clarify the location and functional impact of coronary artery lesions and predict the progression of vascular lesions; further, the embodiments of the present invention locate the branch opening position of the coronary artery based on the vascular type, thereby helping to identify the origin abnormality or stenosis lesions of the coronary artery and helping doctors judge the severity of the lesion and its potential impact on myocardial function; the embodiments of the present invention extract the structural features of the multi-category plaques and analyze the stability of the coronary artery plaques based on the structural features, thereby accurately identifying vulnerable plaques, improving the detection accuracy of vulnerable coronary artery plaques, helping to set up the coronary artery plaque classification detection system, identifying high-risk patients and intervening in advance, reducing the morbidity and mortality of cardiovascular events, achieving accurate diagnosis and treatment of cardiovascular diseases, and improving medical efficiency; further Furthermore, embodiments of the present invention calculate the coronary artery plaque vulnerability index by combining the plaque composition, the radial wall strain value, and the hemodynamic parameters. This allows for early identification of high-risk plaques before plaque rupture and the onset of clinical symptoms, and timely detection of increasing trends in plaque vulnerability, providing the optimal opportunity for clinical intervention. Embodiments of the present invention construct a coronary artery vulnerable plaque scoring mechanism based on the plaque vulnerability index, the plaque composition, and the radial wall strain value. This significantly improves the accuracy and robustness of plaque vulnerability assessment, provides early warning for high-risk patients, and reduces the incidence of cardiovascular events. Furthermore, embodiments of the present invention generate coronary artery vulnerable plaque feedback results by combining the vulnerable plaque scoring mechanism with the plaque classification and detection system. This provides more accurate plaque classification and risk assessment results, enhances diagnostic accuracy, and offers personalized treatment recommendations, assisting physicians in developing precise treatment plans and promoting the intelligent development of cardiovascular disease diagnostic technology. Therefore, the coronary vulnerable plaque detection and analysis system provided by the embodiment of the present invention can improve the accuracy of coronary vulnerable plaque detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 This is a functional module diagram of a coronary artery vulnerable plaque detection and analysis system provided by one embodiment of the present invention;

[0077] Figure 2 A schematic diagram of a flow chart for implementing a method for detecting and analyzing vulnerable coronary artery plaques according to an embodiment of the present invention.

[0078] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0079] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0080] In addition, the step sequence in the following method embodiments is only an example and not a strict limitation.

[0081] In practice, the server-side device deployed in a coronary vulnerable plaque detection and analysis system may be composed of one or more devices. The aforementioned coronary vulnerable plaque detection and analysis system can be implemented as a service instance, a virtual machine, or hardware devices. For example, the coronary vulnerable plaque detection and analysis system can be implemented as a service instance deployed on one or more devices in a cloud node. Simply put, the coronary vulnerable plaque detection and analysis system can be understood as software deployed on a cloud node, providing a coronary vulnerable plaque detection and analysis service to each client. Alternatively, the coronary vulnerable plaque detection and analysis system can be implemented as a virtual machine deployed on one or more devices in a cloud node. Application software for managing each client is installed in the virtual machine. Alternatively, the coronary vulnerable plaque detection and analysis system can be implemented as a server-side device composed of multiple hardware devices of the same or different types, with one or more hardware devices configured to provide a coronary vulnerable plaque detection and analysis service to each client.

[0082] In terms of implementation, a coronary artery vulnerable plaque detection and analysis system and a user terminal are mutually compatible. Specifically, if the coronary artery vulnerable plaque detection and analysis system is an application installed on a cloud service platform, the user terminal is the client that establishes a communication connection with the application; or if the coronary artery vulnerable plaque detection and analysis system is implemented as a website, the user terminal is implemented as a webpage; or if the coronary artery vulnerable plaque detection and analysis system is implemented as a cloud service platform, the user terminal is implemented as a mini-program within an instant messaging application.

[0083] Reference Figure 1 FIG. 1 is a functional module diagram of a coronary vulnerable plaque detection and analysis system provided by an embodiment of the present invention.

[0084] The coronary vulnerable plaque detection and analysis system 100 described in the present invention can be installed in a cloud server. In terms of implementation, it can be implemented as one or more service devices, as an application installed in the cloud (e.g., a coronary vulnerable plaque detection and analysis server, server cluster, etc.), or developed as a website. Depending on the functionality implemented, the coronary vulnerable plaque detection and analysis system 100 includes a data collection module 101, a plaque region identification module 102, a plaque classification and detection module 103, a vulnerable plaque assessment module 104, and a result feedback module 105.

[0085] In an embodiment of the present invention, each of the aforementioned modules can be independently implemented and called upon from other modules in a tracking system based on coronary vulnerable plaque detection and analysis. The term "call" here can be understood as a module being able to connect to multiple modules of a different type and provide corresponding services to these connected modules. In the coronary vulnerable plaque detection and analysis system provided by an embodiment of the present invention, the scope of application of the coronary vulnerable plaque detection and analysis architecture can be adjusted by adding modules and directly calling them, without modifying the program code. This allows for clustered horizontal expansion, enabling the rapid and flexible expansion of the coronary vulnerable plaque detection and analysis system. In practical applications, the aforementioned modules can be deployed within the same device, different devices, or even within virtual devices, such as service instances within a cloud server.

[0086] The following describes various components and specific workflows of a coronary vulnerable plaque detection and analysis system in conjunction with specific embodiments.

[0087] The data collection module 101 is configured to collect multimodal imaging data of the coronary arteries, identify the vascular structure of the coronary arteries based on the multimodal imaging data, determine the vascular type of the coronary arteries based on the vascular structure, and acquire hemodynamic parameters of the coronary arteries based on the vascular structure and the vascular type.

[0088] By collecting multimodal imaging data of the coronary arteries, the embodiments of the present invention can avoid misdiagnosis or missed diagnosis that may occur with single-modality imaging. The coronary arteries refer to the arterial blood vessels that supply blood to the heart. The multimodal imaging data refers to various types of data about the coronary arteries obtained through various imaging technologies or detection methods, such as coronary angiography data, CT vascular imaging, and biomarker detection data.

[0089] Furthermore, the embodiments of the present invention can clarify the location, degree, and range of coronary artery stenosis or obstruction by identifying the vascular structure of the coronary artery based on the multimodal imaging data. The vascular structure refers to the anatomical composition of the coronary artery, including the three-layer structure of the vascular wall (intima, media, and adventitia) as well as the branches and paths of the blood vessels.

[0090] Optionally, the vascular structure identification of the coronary artery based on the multimodal imaging data can be obtained by optical coherence tomography.

[0091] The embodiment of the present invention helps to clarify the location and functional impact of coronary artery lesions by determining the vascular type of the coronary artery based on the vascular structure. The vascular type refers to the classification of blood vessels based on their structure, function, and anatomical characteristics, including arteries, veins, and capillaries.

[0092] As an embodiment of the present invention, determining the vascular type of the coronary artery based on the vascular structure includes: acquiring image data of the coronary artery based on the vascular structure; extracting the vascular centerline and vascular branches of the coronary artery from the image data; identifying the vascular layout of the coronary artery based on the vascular centerline and the vascular branches; calculating the branching angles of the vascular branches and the degree of curvature of the vascular centerline based on the vascular layout; identifying the vascular lumen and vascular outer wall of the coronary artery based on the image data, and measuring the inner lumen diameter and outer wall thickness corresponding to the vascular lumen and the vascular outer wall; extracting multidimensional features of the vascular structure based on the branching angle, the degree of curvature, the inner lumen diameter, and the outer wall thickness; and determining the vascular type of the coronary artery based on the multidimensional features.

[0093] Among them, the image data refers to a two-dimensional or three-dimensional image of the coronary artery obtained by imaging technology, such as generating a high-resolution three-dimensional image of the coronary artery by CT scanning combined with a contrast agent. The blood vessel centerline refers to a line along the central path of the blood vessel lumen, which is used to describe the direction of the main trunk and branches of the blood vessel. The blood vessel branches refer to secondary blood vessels branching from the main blood vessels. The blood vessel layout refers to the overall distribution and connection relationship of the blood vessel centerline and branches. The branching angle refers to the angle between the blood vessel branches. The degree of curvature refers to the degree of curvature of the blood vessel centerline. The blood vessel lumen refers to the space inside the blood vessel for blood flow. The blood vessel outer wall refers to the outer boundary of the blood vessel, including the intima, media and adventitia. The lumen diameter refers to the diameter of the blood vessel lumen. The outer wall thickness refers to the distance between the outer wall and the lumen of the blood vessel. The multidimensional feature refers to a comprehensive feature vector combining multiple geometric and anatomical features.

[0094] Optionally, the lumen diameter measurement corresponding to the vascular lumen can be achieved by measuring the boundary of the vascular lumen, the outer wall thickness corresponding to the vascular outer wall can be determined by measuring the distance between the vascular outer wall and the lumen boundary, the curvature calculation of the vascular centerline based on the vascular layout can be achieved by using the curvature formula of the fitting curve, the branching angle calculation of the vascular branch can be obtained using the vascular radius corresponding to the branch, and the determination of the vascular type of the coronary artery based on the multidimensional features can be achieved through a machine learning algorithm, such as a random forest.

[0095] Furthermore, the embodiments of the present invention can predict the progression of vascular lesions and assess the risk of plaque rupture by collecting the hemodynamic parameters of the coronary artery based on the vascular structure and the vascular type. The hemodynamic parameters refer to physical quantities that describe the flow state of blood in the blood vessels, including blood flow velocity, wall shear force, pressure gradient, flow rate, etc.

[0096] Optionally, the hemodynamic parameters of the coronary artery based on the vascular structure and the vascular type can be acquired through computational fluid simulation, such as constructing a three-dimensional anatomical model using coronary CT or MRI data, combining boundary conditions such as blood pressure and cardiac output, and simulating parameters such as blood flow velocity, pressure and wall shear stress.

[0097] The plaque region identification module 102 is used to locate the branch opening position of the coronary artery based on the blood vessel type, identify the lesion site of the coronary artery according to the branch opening position, and separate multiple types of plaques of the coronary artery based on the lesion site.

[0098] The embodiment of the present invention helps to identify abnormal origins or stenosis of the coronary arteries by locating the branch opening positions of the coronary arteries based on the blood vessel type. The branch opening position refers to the specific location of the starting point of the coronary artery trunk and its branch vessels on the surface of the heart or the aorta root, such as branching from the left coronary artery trunk and descending along the anterior interventricular groove.

[0099] As an embodiment of the present invention, locating the branch opening position of the coronary artery based on the blood vessel type includes: identifying the branch pattern of the coronary artery based on the blood vessel type, and extracting the blood vessel branch points of the coronary artery under the branch pattern; identifying the blood supply organ of the coronary artery, and extracting the organ decomposition characteristics of the blood supply organ; constructing a reference coordinate system of the coronary artery based on the organ decomposition characteristics; extracting the variant blood vessel characteristics of the coronary artery according to the blood vessel type; generating a spatial mapping network of the blood vessel type and the blood vessel branch points based on the variant blood vessel characteristics and the branch pattern; and locating the branch opening position of the coronary artery by combining the spatial mapping network and the reference coordinate system.

[0100] Among them, the branching pattern refers to the distribution and direction of vascular branches in the anatomical structure of the coronary artery, including the branches of the left coronary artery and the right coronary artery and their distribution areas; the vascular branch point refers to the location where the main trunk or branch vessels of the coronary artery bifurcate; the blood supply organ refers to the heart organ supplied by the coronary artery; the organ decomposition characteristics refer to the anatomical characteristics of the heart including the apex, base of the heart, ventricular wall thickness, heart cavity size, etc.; the reference coordinate system refers to a coordinate system established based on the anatomical characteristics of the blood supply organ; the variant vascular characteristics refer to the differences between the blood vessels and the normal anatomical structure in terms of origin, course, branch distribution, morphology and function; the spatial mapping network refers to a network model that maps the branch points and vascular types of the coronary arteries to specific spatial positions in the reference coordinate system.

[0101] Optionally, the extraction of the vascular branch points of the coronary artery in the branching pattern based on the vascular type can be achieved by combining image processing and convolutional neural networks. The construction of the reference coordinate system of the coronary artery based on the organ decomposition characteristics can be achieved using a global coordinate system, such as an LPS coordinate system. The extraction of the variant vascular features of the coronary artery based on the vascular type can be obtained through multi-slice spiral CT angiography. The generation of the spatial mapping network of the vascular type and the vascular branch points based on the variant vascular features and the branching pattern can be achieved using a graph neural network.

[0102] Furthermore, the embodiments of the present invention can help doctors determine the severity of the lesion and its potential impact on myocardial function and guide interventional treatment by identifying the lesion site of the coronary artery based on the location of the branch opening. The lesion site refers to the specific location in the coronary artery where stenosis, blockage or other pathological changes occur.

[0103] Optionally, the identification of the lesion site of the coronary artery based on the branch opening position can be determined by CT angiography technology.

[0104] The embodiments of the present invention help doctors more intuitively understand the vulnerability of plaques by separating multiple types of plaques in the coronary artery based on the lesion site. The multiple types of plaques refer to various types of deposits with different characteristics, components and morphologies present on the coronary artery wall, such as lipid plaques, fibrous plaques, calcified plaques, etc.

[0105] As an embodiment of the present invention, the method of extracting multiple categories of plaques from the coronary artery based on the lesion site includes: collecting a multimodal vascular image of the coronary artery based on the lesion site; identifying a decomposition structure of the coronary artery and extracting pathological features of the lesion site based on the multimodal vascular image; extracting a deposit area of ​​the coronary artery from the multimodal vascular image based on the decomposition structure and the pathological features; identifying the grayscale distribution and plaque geometry of the deposit area; extracting plaque deposits from the coronary artery based on the grayscale distribution and the plaque geometry; extracting depth texture features of the plaque deposits in the deposit area; and extracting multiple categories of plaques from the coronary artery based on a combination of the grayscale distribution, the plaque geometry, and the depth texture features.

[0106] Among them, the multimodal vascular image refers to a coronary artery vascular image obtained by multiple imaging technologies, the decomposition structure refers to the anatomical structure of the coronary artery, such as the trunk, branches, openings, etc., the pathological characteristics refer to the pathological characteristics of the coronary artery lesion site, including the composition, morphology, size and distribution of the plaque, the deposit area refers to the area where the coronary artery plaque is located identified in the multimodal vascular image, the grayscale distribution refers to the distribution of pixel grayscale values ​​in the image, the plaque geometry refers to the shape, size and distribution characteristics of the plaque, such as the area, perimeter, shape factor and reconstruction index of the plaque, the plaque deposit refers to the deposit formed on the blood vessel wall, the main components of which include cholesterol, lipids, calcium salts and fibrous tissue, etc., and the deep texture features refer to image texture features extracted by deep learning algorithms (such as convolutional neural networks), which can capture subtle texture changes in the image.

[0107] Optionally, based on the multimodal vascular image, the pathological feature extraction of the lesion site can be obtained through a vascular cross-section, and based on the decomposition structure and the pathological features, the deposit area of ​​the coronary artery can be extracted from the multimodal vascular image by automatic segmentation using a CAB-U-Net deep learning model. The grayscale distribution of the deposit area can be identified by a grayscale histogram, and the geometric morphology of the plaque in the deposit area can be determined using a shape descriptor, such as a Hu moment descriptor. The deep texture feature extraction of the plaque deposit in the deposit area can be achieved by a convolutional neural network or an autoencoder.

[0108] The plaque classification detection module 103 is used to extract the structural features of the multi-category plaques, analyze the plaque stability of the coronary artery based on the structural features, identify vulnerable plaques and stable plaques in the coronary artery based on the plaque stability, and set up the plaque classification detection system of the coronary artery based on the vulnerable plaques and the stable plaques.

[0109] The embodiments of the present invention can accurately identify vulnerable plaques and improve the detection accuracy of vulnerable coronary artery plaques by extracting the structural features of the multiple categories of plaques and analyzing the plaque stability of the coronary artery based on the structural features. The structural features refer to the characteristics of the morphology, composition and internal organization of atherosclerotic plaques, such as inflammatory cell infiltration characteristics and fibrous cap thickness. The plaque stability refers to the ability of the plaque to maintain structural integrity within the blood vessel wall and not easily rupture or fall off.

[0110] Optionally, the structural feature extraction of the multi-category plaques can be obtained by a K-means clustering method.

[0111] As an embodiment of the present invention, analyzing the stability of the coronary artery plaque based on the structural characteristics includes: analyzing the structural complexity of the multi-category plaques based on the structural characteristics, and identifying characteristic elements corresponding to the multi-category plaques; performing normalization processing on the characteristic elements to obtain normalized characteristic elements; calculating the probability distribution ratio of the normalized characteristic elements; calculating the characteristic entropy value corresponding to the normalized characteristic element based on the probability distribution ratio; and analyzing the stability of the coronary artery plaque based on the structural complexity and the characteristic entropy value.

[0112] Among them, the structural complexity refers to the diversity and heterogeneity of the internal structure of the plaque. For example, if a plaque contains a thin fibrous cap, a large lipid core, microcalcifications and high inflammatory activity, its structural complexity is relatively high. On the contrary, if a plaque has a thicker fibrous cap, a smaller lipid core and evenly distributed calcifications, its structural complexity is relatively low. The characteristic elements refer to specific parameters used to describe the plaque structure, such as the fibrous cap and lipid core. The normalization processing refers to the process of converting data into a unified numerical range (usually [0,1]). The normalized characteristic elements refer to the characteristic elements after normalization processing. The probability distribution ratio refers to the probability distribution of each normalized characteristic element in all samples. The characteristic entropy value refers to a numerical value calculated based on the probability distribution ratio and used to measure the complexity and uncertainty of the characteristic elements.

[0113] Optionally, based on the structural features, the structural complexity analysis of the multi-category plaques can be achieved through a feature fusion method, the normalization processing of the characteristic elements can be achieved using the Min-Max normalization formula, and the probability distribution ratio calculation of the normalized characteristic elements can be obtained by statistically analyzing the distribution of each normalized characteristic value in the data.

[0114] In an optional embodiment of the present invention, based on the probability distribution ratio, the characteristic entropy value corresponding to the normalized characteristic element is calculated using the following formula:

[0115]

[0116] Among them, H(X) represents the characteristic entropy value corresponding to the normalized characteristic element, P(x j ) represents the probability distribution ratio corresponding to the normalized feature element, x j represents the jth normalized value corresponding to the normalized feature element x, m represents the number of normalized values ​​corresponding to the normalized feature element, and j represents the index number of the normalized value of the normalized feature element x.

[0117] Furthermore, the embodiments of the present invention can identify vulnerable plaques and stable plaques in the coronary artery based on the stability of the plaque, thereby identifying vulnerable plaques at an early stage and implementing targeted intervention, significantly reducing the incidence of cardiovascular events while avoiding excessive treatment of stable plaques. The vulnerable plaque refers to an atherosclerotic plaque with an unstable structure that is prone to rupture or erosion, and the stable plaque refers to a plaque surrounded by complete fibrous connective tissue and vascular endothelium, and is usually composed of a lipid core surrounded by a fibrous cap.

[0118] As an embodiment of the present invention, the identifying of vulnerable plaques and stable plaques in the coronary artery based on the plaque stability includes: acquiring medical imaging data of the coronary artery, and extracting the vascular wall and plaque deposition area of ​​the coronary artery from the medical imaging data; extracting morphological characteristics, component characteristics, and hemodynamic characteristics of the plaque deposition area, and extracting geometric characteristics of the vascular wall; scoring the stability of different plaques in the plaque deposition area based on the morphological characteristics, the component characteristics, the hemodynamic characteristics, and the geometric characteristics to obtain a plaque scoring result; setting a stability threshold for the different plaques according to the plaque stability; and identifying vulnerable plaques and stable plaques in the coronary artery based on the stability threshold and the plaque scoring result.

[0119] The medical imaging data refers to image data of coronary arteries acquired using medical imaging technology; the vascular wall refers to the coronary artery wall structure, including the intima, media, and adventitia; the plaque deposition area refers to the area on the coronary artery wall where plaques such as lipids, calcifications, and fibrous tissue are deposited; the morphological characteristics refer to geometric features such as shape, size, and surface roughness of the plaque deposition area; the compositional characteristics refer to the components of the plaque deposition area, such as the lipid core, calcified area, and fibrous cap; the hemodynamic characteristics refer to hemodynamic parameters around the plaque area, such as wall shear stress, blood flow velocity, and pressure gradient; the geometric characteristics refer to geometric properties of the vascular wall, such as vascular diameter, curvature, and wall thickness; the different plaques refer to different types of plaques in the plaque deposition area, such as calcified plaques, lipid plaques, and fibrous plaques; the plaque scoring result refers to the scoring result of each plaque using a plaque stability assessment model; and the stability threshold refers to the scoring threshold used to distinguish between vulnerable plaques and stable plaques.

[0120] Optionally, the stability scores of different plaques in the plaque deposition area based on the morphological characteristics, the compositional characteristics, the hemodynamic characteristics and the geometric characteristics can be achieved using a random forest model, and the stability threshold settings of the different plaques according to the plaque stability can be determined by clinical data.

[0121] The embodiment of the present invention sets up the coronary artery plaque classification detection system based on the vulnerable plaque and the stable plaque, thereby identifying high-risk patients and intervening in advance, reducing the morbidity and mortality of cardiovascular events, and achieving accurate diagnosis and treatment of cardiovascular diseases. At the same time, it can rationally allocate medical resources, give priority to treating high-risk patients, and improve medical efficiency. The plaque classification detection system refers to a system that can identify, analyze and classify plaques present in the coronary arteries.

[0122] As an embodiment of the present invention, the coronary artery plaque classification detection system is set up based on the vulnerable plaque and the stable plaque, including: extracting the classified plaque characteristics of the coronary artery based on the vulnerable plaque and the stable plaque; detecting blood markers of the coronary artery, and identifying the inflammatory marker level of the coronary artery based on the blood marker; calculating the correlation coefficient between the inflammatory marker level and the classified plaque characteristics; analyzing the correlation pattern between the inflammatory marker level and the classified plaque characteristics based on the correlation coefficient; and setting up the coronary artery plaque classification detection system based on the correlation pattern.

[0123] Among them, the classified plaque characteristics refer to specific imaging or pathological characteristics used to distinguish the stability and vulnerability of coronary artery plaques, such as fibrous cap thickness characteristics, lipid volume characteristics, etc.; the blood markers refer to biological molecules that can be detected in the blood, such as C-reactive protein, interleukin-6, etc.; the inflammatory marker levels refer to the concentrations of inflammatory markers quantitatively analyzed by blood tests, reflecting the degree of plaque inflammatory activity; the correlation coefficients refer to the strength of the correlation between the inflammatory marker levels and the plaque characteristics; the association patterns refer to the relationship patterns between the inflammatory marker levels and the plaque characteristics obtained through statistical analysis, such as high IL-6 levels are positively correlated with low-density core plaques, and low IL-6 levels are positively correlated with calcified plaques.

[0124] Optionally, the detection of blood markers in the coronary arteries can be achieved through Luminex technology, the identification of the inflammatory marker levels in the coronary arteries based on the blood markers can be obtained by immunohistochemistry, and the analysis of the association pattern between the inflammatory marker levels and the classified plaque characteristics based on the correlation coefficient can be achieved by principal component analysis.

[0125] The vulnerable plaque assessment module 104 is configured to extract the plaque composition of the coronary artery from the structural features, measure the radial wall strain value of the coronary artery, calculate the plaque vulnerability index of the coronary artery based on the plaque composition, the radial wall strain value, and the hemodynamic parameters, and construct a vulnerable plaque scoring mechanism for the coronary artery based on the plaque vulnerability index, the plaque composition, and the radial wall strain value.

[0126] The embodiments of the present invention can help identify high-risk vulnerable plaques and predict the risk of cardiovascular events by extracting the plaque components of the coronary artery from the structural features and measuring the radial wall strain value of the coronary artery. The plaque components refer to the internal structure and substances that constitute the coronary atherosclerotic plaque, such as lipid components, fiber components, necrotic core, etc. The radial wall strain value refers to the deformation coefficient of the blood vessel wall in the radial direction during the cardiac cycle.

[0127] Optionally, the radial wall strain value measurement of the coronary artery can be obtained by cardiac imaging technology, such as using speckle tracking technology to track the movement trajectory of acoustic speckles in the myocardium during the cardiac cycle, calculate the strain of each point in the myocardium in different directions, and thus obtain the radial wall strain.

[0128] Furthermore, embodiments of the present invention calculate the coronary artery plaque vulnerability index by combining the plaque composition, the radial wall strain value, and the hemodynamic parameters. This allows for early identification of high-risk plaques before the plaque ruptures and clinical symptoms appear, and timely detection of increasing trends in plaque vulnerability, providing the best opportunity for clinical intervention. The plaque vulnerability index is a quantitative indicator used to comprehensively assess whether a coronary artery plaque is prone to rupture and trigger cardiovascular events.

[0129] As an embodiment of the present invention, the calculation of the plaque vulnerability index of the coronary artery by combining the plaque composition, the radial wall strain value and the hemodynamic parameters includes: extracting the lipid core ratio and fibrous cap thickness of the coronary artery based on the plaque composition; analyzing the shear stress, blood flow velocity and blood flow direction of the coronary artery based on the hemodynamic parameters; identifying the vascular oscillation shear index of the coronary artery based on the shear stress, the blood flow velocity and the blood flow direction; and calculating the plaque vulnerability index of the coronary artery based on the lipid core ratio, the fibrous cap thickness, the shear stress and the vascular oscillation shear index.

[0130] Among them, the lipid core ratio refers to the proportion of the lipid core in the plaque to the entire plaque volume, the fibrous cap thickness refers to the thickness of the fibrous tissue covering the lipid core, the shear stress refers to the tangential force exerted by the blood flow on the blood vessel wall, the blood flow velocity refers to the flow velocity of the blood in the blood vessel, the blood flow direction refers to the flow direction of the blood in the blood vessel, and the vascular oscillation shear index refers to an indicator used to quantify the frequency of blood flow direction changes and the degree of shear stress oscillation.

[0131] Optionally, the shear stress analysis of the coronary artery based on the hemodynamic parameters can be obtained by Doppler ultrasound, and the extraction of the lipid core ratio and fibrous cap thickness of the coronary artery based on the plaque composition can be achieved by intravascular ultrasound.

[0132] In an optional embodiment of the present invention, the coronary artery plaque vulnerability index is calculated using the following formula in combination with the plaque composition, the radial wall strain value, and the hemodynamic parameter:

[0133] I=w1·L+w2·(1-C)+w3·S+w4·(1-B)+w5·O

[0134] Among them, I represents the coronary artery plaque vulnerability index, w1·L represents the contribution value of the lipid core ratio in the plaque component, w2·(1-C) represents the contribution value corresponding to the fibrous cap thickness in the plaque component, w3·S represents the contribution value corresponding to the radial wall strain value, w4·(1-B) represents the contribution value corresponding to the shear stress in the hemodynamic parameters, and w5·O represents the contribution value corresponding to the vascular oscillation shear index.

[0135] It should be explained that the thinner the fibrous cap, the more vulnerable the plaque, so 1-C is used to represent the contribution of fibrous cap thickness to the plaque vulnerability index, and low shear stress is related to plaque vulnerability, so 1-B is used to represent the contribution of shear stress to the plaque vulnerability index.

[0136] The embodiments of the present invention construct a vulnerable plaque scoring mechanism for the coronary artery based on the plaque vulnerability index, the plaque composition, and the radial wall strain value, thereby significantly improving the accuracy and robustness of plaque vulnerability assessment, providing early warning for high-risk patients, and reducing the incidence of cardiovascular events. The plaque vulnerability scoring mechanism refers to a system for assessing the risk level of adverse events such as rupture and thrombosis in coronary atherosclerotic plaques.

[0137] As an embodiment of the present invention, the vulnerable plaque scoring mechanism for the coronary artery is constructed based on the plaque vulnerability index, the plaque composition and the radial wall strain value, including: defining a plaque vulnerability evaluation index for the coronary artery based on the plaque vulnerability index, the plaque composition and the radial wall strain value; calculating a plaque stability score for the coronary artery based on the plaque vulnerability evaluation index; identifying a numerical range for the plaque stability score and defining a grading threshold for the plaque stability score based on the numerical range; setting a plaque damage grade for the coronary artery based on the grading threshold; extracting plaque features corresponding to the plaque vulnerability evaluation index and identifying a plaque category corresponding to the plaque features; creating a result interpretation of the plaque score based on the plaque features, the plaque category and the plaque damage grade; and generating a plaque vulnerability scoring mechanism for the coronary artery based on the plaque stability score and the result interpretation.

[0138] Among them, the plaque vulnerability evaluation index refers to a quantitative index for comprehensively evaluating the stability and rupture risk of coronary artery plaques, such as the lipid core ratio. The plaque stability score refers to the plaque stability value calculated by comprehensively evaluating the plaque vulnerability evaluation index. The numerical interval refers to the range of the plaque stability score, usually from 0 to 1. The grading threshold refers to the critical value used to divide the numerical interval, for example, 0.33 and 0.67 can be used as grading thresholds. The plaque damage grade refers to the plaque risk grade divided according to the grading threshold, which is usually divided into Low damage, moderate damage and high damage levels, for example, low risk interval: PVI < 0.5, moderate risk interval: 0.5 ≤ PVI < 1.0, high risk interval: PVI ≥ 1.0, the result interpretation refers to the specific interpretation of the plaque scoring results, the plaque characteristics refer to the key attributes extracted from the plaque vulnerability evaluation indicators, such as lipid core ratio, fibrous cap thickness, calcification components, inflammatory cell infiltration, etc., the plaque category refers to the plaque type defined according to the plaque characteristics, usually including calcified plaque, lipid plaque, fibrous plaque and mixed plaque.

[0139] Optionally, the calculation of the coronary artery plaque stability score based on the plaque vulnerability evaluation index can be achieved using a weighted summation formula, and the interpretation of the plaque score result based on the plaque characteristics, the plaque category and the plaque damage level can be obtained through visualization tools, such as icon tools.

[0140] The embodiment of the present invention defines the coronary artery plaque vulnerability evaluation index based on the plaque vulnerability index, the plaque composition, and the radial wall strain value, which can more comprehensively reflect the stability of the plaque, thereby improving the accuracy of vulnerable plaque detection and facilitating doctors to intuitively assess the risk level of the plaque. The plaque vulnerability evaluation index refers to a quantitative indicator for comprehensively evaluating the stability and rupture risk of coronary artery plaques, such as the lipid core ratio.

[0141] The result feedback module 105 is used to generate a vulnerable plaque feedback result of the coronary artery by combining the vulnerable plaque scoring mechanism and the plaque classification detection system.

[0142] The embodiments of the present invention combine the vulnerable plaque scoring mechanism and the plaque classification detection system to generate the vulnerable plaque feedback results of the coronary artery, which can provide more accurate plaque classification and risk assessment results, improve the accuracy of diagnosis, and provide personalized treatment recommendations, assisting doctors in formulating accurate treatment plans, and promoting the intelligent development of cardiovascular disease diagnosis technology. The vulnerable plaque feedback results refer to the conclusive opinions obtained after a comprehensive analysis of the characteristics, risks and potential impacts of vulnerable plaques in the coronary artery, such as recommending patients to improve their lifestyles, such as quitting smoking, controlling diet, increasing exercise, etc., to reduce the risk of plaque progression.

[0143] As an embodiment of the present invention, the vulnerable plaque scoring mechanism and the plaque classification detection system are combined to generate vulnerable plaque feedback results of the coronary artery, including: identifying the classified plaques of the coronary artery based on the plaque classification detection system; collecting serial images of the classified plaques, and extracting the historical change trend of the classified plaques from the serial images; generating a disease progression report of the classified plaques based on the historical change trend; outputting the plaque score of the classified plaque based on the vulnerable plaque scoring mechanism; generating personalized feedback opinions on the classified plaques based on the plaque score and the disease progression report; and outputting the vulnerable plaque feedback results of the coronary artery based on the disease progression report and the personalized feedback opinions.

[0144] Among them, the plaque classification refers to classifying coronary artery plaques into different types of plaques, such as vulnerable plaques and stable plaques, based on their morphology, composition and stability characteristics. The serial image refers to an image sequence obtained by performing multiple, continuous imaging examinations on coronary artery plaques. The historical change trend refers to extracting changes in the morphology, composition and size of coronary artery plaques at different time points by analyzing serial images, such as increase in plaque volume, thinning of the fibrous cap or expansion of the lipid core. The disease progression report refers to a document generated based on the historical change trend, which records in detail the dynamic changes of coronary artery plaques, including information such as changes in plaque stability and increase or decrease in vulnerability. The plaque score refers to a quantitative assessment result of the risk level of coronary artery plaques. The personalized feedback refers to targeted suggestions provided to patients or doctors based on the plaque score and disease progression report.

[0145] Optionally, the historical change trend of the classified plaques extracted from the serial images can be obtained by comparing the image data at different time points, and the generation of personalized feedback opinions on the classified plaques based on the plaque score and the disease progression report can be achieved through an online platform or mobile application.

[0146] Compared with the problems described in the background technology, the embodiments of the present invention collect multimodal imaging data of the coronary arteries and identify the vascular structure and vascular type of the coronary arteries, thereby avoiding misdiagnosis or missed diagnosis that may occur in single-modal imaging, helping to clarify the location and functional impact of coronary artery lesions and predict the progression of vascular lesions; further, the embodiments of the present invention locate the branch opening position of the coronary artery based on the vascular type, thereby helping to identify the origin abnormality or stenosis lesions of the coronary artery and helping doctors judge the severity of the lesion and its potential impact on myocardial function; the embodiments of the present invention extract the structural features of the multi-category plaques and analyze the stability of the coronary artery plaques based on the structural features, thereby accurately identifying vulnerable plaques, improving the detection accuracy of vulnerable coronary artery plaques, helping to set up the coronary artery plaque classification detection system, identifying high-risk patients and intervening in advance, reducing the morbidity and mortality of cardiovascular events, achieving accurate diagnosis and treatment of cardiovascular diseases, and improving medical efficiency; further Furthermore, embodiments of the present invention calculate the coronary artery plaque vulnerability index by combining the plaque composition, the radial wall strain value, and the hemodynamic parameters. This allows for early identification of high-risk plaques before plaque rupture and the onset of clinical symptoms, and timely detection of increasing trends in plaque vulnerability, providing the optimal opportunity for clinical intervention. Embodiments of the present invention construct a coronary artery vulnerable plaque scoring mechanism based on the plaque vulnerability index, the plaque composition, and the radial wall strain value. This significantly improves the accuracy and robustness of plaque vulnerability assessment, provides early warning for high-risk patients, and reduces the incidence of cardiovascular events. Furthermore, embodiments of the present invention generate coronary artery vulnerable plaque feedback results by combining the vulnerable plaque scoring mechanism with the plaque classification and detection system. This provides more accurate plaque classification and risk assessment results, enhances diagnostic accuracy, and offers personalized treatment recommendations, assisting physicians in developing precise treatment plans and promoting the intelligent development of cardiovascular disease diagnostic technology. Therefore, the coronary vulnerable plaque detection and analysis system provided by the embodiment of the present invention can improve the accuracy of coronary vulnerable plaque detection.

[0147] like Figure 2 FIG. 1 is a flow chart of a method for detecting and analyzing vulnerable coronary artery plaques according to an embodiment of the present invention.

[0148] In this embodiment, the method for detecting and analyzing vulnerable coronary artery plaques includes:

[0149] collecting multimodal imaging data of the coronary arteries, identifying a vascular structure of the coronary arteries based on the multimodal imaging data, determining a vascular type of the coronary arteries based on the vascular structure, and acquiring hemodynamic parameters of the coronary arteries based on the vascular structure and the vascular type;

[0150] Locating the branch opening position of the coronary artery based on the blood vessel type, identifying the lesion site of the coronary artery according to the branch opening position, and separating multiple types of plaques in the coronary artery based on the lesion site;

[0151] extracting structural features of the multiple categories of plaques, and analyzing the plaque stability of the coronary artery based on the structural features, identifying vulnerable plaques and stable plaques in the coronary artery based on the plaque stability, and setting a plaque classification detection system for the coronary artery based on the vulnerable plaques and the stable plaques;

[0152] extracting the coronary artery plaque composition from the structural features, measuring the radial wall strain value of the coronary artery, calculating the coronary artery plaque vulnerability index based on the plaque composition, the radial wall strain value, and the hemodynamic parameter, and constructing a coronary artery vulnerable plaque scoring mechanism based on the plaque vulnerability index, the plaque composition, and the radial wall strain value;

[0153] The vulnerable plaque scoring mechanism and the plaque classification detection system are combined to generate vulnerable plaque feedback results of the coronary artery.

[0154] In the several embodiments provided by the present invention, it should be understood that the provided systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the module division is merely a logical function division, and actual implementation may employ other division methods.

[0155] In addition, the functional modules in various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or hardware plus software functional modules.

[0156] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A coronary artery vulnerable plaque detection and analysis system, characterized in that: The coronary vulnerable plaque detection and analysis system comprises: a data collection module, a plaque area recognition module, a plaque classification detection module, a vulnerable plaque assessment module and a result feedback module; The data collection module is configured to collect multimodal imaging data of the coronary arteries, identify the vascular structure of the coronary arteries based on the multimodal imaging data, determine the vascular type of the coronary arteries based on the vascular structure, and acquire hemodynamic parameters of the coronary arteries based on the vascular structure and the vascular type; The plaque region identification module is configured to locate the branch opening positions of the coronary artery based on the blood vessel type, identify the lesion site of the coronary artery based on the branch opening positions, and separate out multiple types of plaques in the coronary artery based on the lesion site; The plaque classification and detection module is configured to extract structural features of the multiple types of plaques, analyze the plaque stability of the coronary artery based on the structural features, identify vulnerable plaques and stable plaques in the coronary artery based on the plaque stability, and establish a plaque classification and detection system for the coronary artery based on the vulnerable plaques and the stable plaques; The vulnerable plaque assessment module is configured to extract the plaque composition of the coronary artery from the structural features, measure the radial wall strain value of the coronary artery, calculate the plaque vulnerability index of the coronary artery based on the plaque composition, the radial wall strain value, and the hemodynamic parameter, and construct a vulnerable plaque scoring mechanism for the coronary artery based on the plaque vulnerability index, the plaque composition, and the radial wall strain value; The result feedback module is used to generate vulnerable plaque feedback results of the coronary artery by combining the vulnerable plaque scoring mechanism and the plaque classification detection system.

2. A coronary artery vulnerable plaque detection and analysis system according to claim 1, characterized in that: Determining the blood vessel type of the coronary artery according to the blood vessel structure includes: collecting image data of the coronary artery according to the vascular structure; extracting a vascular centerline and vascular branches of the coronary artery from the image data; identifying a vascular layout of the coronary artery based on the vascular centerline and the vascular branches; Calculating the branching angles of the blood vessel branches and the curvature of the blood vessel centerline according to the blood vessel layout; Based on the image data, identifying the vascular lumen and vascular outer wall of the coronary artery, and measuring the corresponding inner lumen diameter and outer wall thickness of the vascular lumen and the vascular outer wall; extracting multi-dimensional features of the blood vessel structure by combining the branch angle, the degree of curvature, the inner cavity diameter, and the outer wall thickness; The vessel type of the coronary artery is determined according to the multidimensional features.

3. The coronary artery vulnerable plaque detection and analysis system according to claim 1, characterized in that: The locating the branch opening position of the coronary artery based on the blood vessel type includes: Based on the blood vessel type, identifying a branching pattern of the coronary artery, and extracting blood vessel branch points of the coronary artery in the branching pattern; identifying the blood supply organ of the coronary artery and extracting organ decomposition features of the blood supply organ; constructing a reference coordinate system of the coronary artery based on the organ decomposition characteristics; Identifying variant vascular features of the coronary artery according to the vascular type; generating a spatial mapping network of the blood vessel types and the blood vessel branch points based on the variant blood vessel characteristics and the branching pattern; The branch opening position of the coronary artery is located by combining the spatial mapping network and the reference coordinate system.

4. The coronary artery vulnerable plaque detection and analysis system according to claim 1, characterized in that: The method of separating out multiple types of plaques in the coronary artery based on the lesion site includes: collecting a multimodal vascular image of the coronary artery based on the lesion location; identifying the decomposed structure of the coronary artery and extracting the pathological features of the lesion site based on the multimodal vascular image; extracting the deposit region of the coronary artery from the multimodal vascular image based on the decomposition structure and the pathological features; Identify the grayscale distribution and patch geometry of the sediment area; Separating the coronary artery plaque deposits according to the grayscale distribution and the plaque geometry; Extracting depth texture features of the patch sediment in the sediment area; Combining the grayscale distribution, the geometric shape of the plaque and the depth texture characteristics, multiple categories of plaques in the coronary artery are separated.

5. The coronary vulnerable plaque detection and analysis system according to claim 1, characterized in that: Analyzing the stability of the coronary artery plaque according to the structural characteristics includes: Analyzing the structural complexity of the multi-category plaques according to the structural characteristics, and identifying characteristic elements corresponding to the multi-category plaques; performing normalization processing on the characteristic elements to obtain normalized characteristic elements; Calculating the probability distribution ratio of the normalized feature elements; Based on the probability distribution ratio, calculating the characteristic entropy value corresponding to the normalized characteristic element; The stability of the coronary artery plaque is analyzed according to the structural complexity and the characteristic entropy value.

6. The coronary vulnerable plaque detection and analysis system according to claim 1, characterized in that: The identifying of vulnerable plaques and stable plaques in the coronary artery based on the plaque stability comprises: collecting medical imaging data of the coronary artery, and extracting the vessel wall and plaque deposition area of ​​the coronary artery from the medical imaging data; extracting morphological features, compositional features, and hemodynamic features of the plaque deposition area, and extracting geometric features of the blood vessel wall; Based on the morphological characteristics, the component characteristics, the hemodynamic characteristics, and the geometric characteristics, a stability score is performed on different plaques in the plaque deposition area to obtain a plaque scoring result; According to the plaque stability, setting the stability thresholds of the different plaques; Based on the stability threshold and the plaque scoring result, vulnerable plaques and stable plaques in the coronary artery are identified.

7. The coronary artery vulnerable plaque detection and analysis system according to claim 1, characterized in that: The coronary artery plaque classification and detection system is set up according to the vulnerable plaque and the stable plaque, including: extracting classified plaque features of the coronary artery according to the vulnerable plaque and the stable plaque; detecting blood markers of the coronary arteries, and identifying the levels of inflammatory markers of the coronary arteries based on the blood markers; calculating the correlation coefficient between the inflammatory marker level and the classified plaque characteristics; analyzing the correlation pattern between the inflammatory marker level and the classified plaque characteristics according to the correlation coefficient; Based on the association pattern, a coronary artery plaque classification and detection system is set up.

8. The coronary vulnerable plaque detection and analysis system according to claim 1, characterized in that: Calculating the coronary artery plaque vulnerability index by combining the plaque composition, the radial wall strain value, and the hemodynamic parameter includes: extracting the lipid core ratio and fibrous cap thickness of the coronary artery based on the plaque composition; Analyzing the shear stress, blood flow velocity, and blood flow direction of the coronary artery based on the hemodynamic parameters; identifying a vascular oscillation shear index of the coronary artery based on the shear stress, the blood flow velocity, and the blood flow direction; The coronary artery plaque vulnerability index is calculated according to the lipid core ratio, the fibrous cap thickness, the shear stress, and the vascular oscillation shear index using the following formula: I=w1·L+w2·(1-C)+w3·S+w4·(1-B)+w5·O Among them, I represents the coronary artery plaque vulnerability index, w1·L represents the contribution value of the lipid core ratio in the plaque component, w2·(1-C) represents the contribution value corresponding to the fibrous cap thickness in the plaque component, w3·S represents the contribution value corresponding to the radial wall strain value, w4·(1-B) represents the contribution value corresponding to the shear stress in the hemodynamic parameters, and w5·O represents the contribution value corresponding to the vascular oscillation shear index.

9. The coronary artery vulnerable plaque detection and analysis system according to claim 1, characterized in that: The constructing of a vulnerable plaque scoring mechanism for the coronary artery based on the plaque vulnerability index, the plaque composition, and the radial wall strain value includes: defining a plaque vulnerability evaluation index of the coronary artery according to the plaque vulnerability index, the plaque composition, and the radial wall strain value; Calculating the coronary artery plaque stability score based on the plaque vulnerability evaluation index; Identifying a numerical range of the plaque stability score, and defining a grading threshold of the plaque stability score based on the numerical range; Setting the plaque damage grade of the coronary artery based on the grade threshold; extracting plaque features corresponding to the plaque vulnerability evaluation index, and identifying the plaque category corresponding to the plaque features; Creating a result interpretation of the plaque score according to the plaque characteristics, the plaque category, and the plaque damage grade; Based on the plaque stability score and the result interpretation, a plaque vulnerability score mechanism for the coronary artery is generated.

10. The coronary vulnerable plaque detection and analysis system according to claim 1, characterized in that: The generating of the vulnerable plaque feedback result of the coronary artery by combining the vulnerable plaque scoring mechanism and the plaque classification detection system includes: Based on the plaque classification detection system, identifying the classified plaques of the coronary artery; collecting sequential images of the classified plaques, and extracting historical change trends of the classified plaques from the sequential images; generating a disease progression report of the classified plaque according to the historical change trend; outputting a plaque score of the classified plaque based on the vulnerable plaque scoring mechanism; generating personalized feedback on the classified plaques based on the plaque score and the disease progression report; Based on the disease progression report and the personalized feedback, a feedback result of the vulnerable plaque of the coronary artery is output.

Images