Angiocardiography image intelligent classification method and system

Abstract

The invention relates to the technical field of image processing, in particular to an intelligent classification method and system for angiocardiography images, and the method comprises the steps: determining a blood vessel segment corresponding to each branch through each branch point after angiocardiography image blood vessel skeletonization; determining a narrow area and a narrow degree by using the blood vessel inner diameter deviation in the target blood vessel section, and further determining an initial abnormal index; feature registration is carried out on the multiple frames of angiocardiography images to obtain a registered target blood vessel section, and the stenosis relieving degree of the registered target blood vessel section is determined according to the initial abnormal index change of the registered target blood vessel section before and after the contrast agent flows into the registered target blood vessel section; determining blood vessel improvement response degrees of upstream and downstream branches of the registered target blood vessel section after stenosis relief, determining a target stenosis index of the registered target blood vessel section by combining the stenosis relief degree, and performing classified early warning based on the target stenosis index. Through the technical scheme of the invention, more accurate quantitative expression of the vascular stenosis and refined classification of the vascular stenosis state are realized.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, specifically to an intelligent classification method and system for cardiovascular angiography images. Background Technology

[0002] Cardiovascular angiography, a technique that uses imaging equipment to visualize the morphology of blood vessels after contrast agents are injected, can accurately reveal whether there are lesions such as stenosis, occlusion, or abnormal dilation in the coronary arteries. It plays an irreplaceable role in helping doctors diagnose cardiovascular diseases. When patients experience symptoms such as chest tightness or chest pain, cardiovascular angiography can visually show the location of vascular stenosis, providing crucial data for determining whether stent implantation or bypass surgery is necessary. From early screening for occult vascular lesions to postoperative assessment of recanalization outcomes, cardiovascular angiography continuously provides scientific support for the diagnosis and treatment of cardiovascular diseases and is an important technological guarantee in modern cardiovascular medicine.

[0003] The degree of vascular stenosis directly affects the normal flow of blood within the vessels, leading not only to increased blood flow resistance but also potentially inducing vascular-related diseases. Scientific classification based on the degree of vascular stenosis can facilitate clinicians in quickly assessing the severity of the condition. However, current methods analyze the stenosis of a single vessel in isolation, failing to comprehensively reflect the true state of blood flow within the vessel segment, resulting in low precision in the segmentation of vascular networks in imaging. Summary of the Invention

[0004] To address the technical problem of low precision in vascular network segmentation in existing solutions for cardiovascular imaging, the present invention aims to provide an intelligent classification method and system for cardiovascular angiography images. The specific technical solution adopted is as follows: This invention provides an intelligent classification method for cardiovascular angiography images, the method comprising: The vascular segment corresponding to each branch is determined by the branch points of the vascular skeletonized image in the cardiovascular angiography. The stenotic region and its degree of stenosis are determined by the deviation of the inner diameter of the target vascular segment, and the initial abnormality index of the target vascular segment is determined by each degree of stenosis. The target vessel segment was obtained by feature registration of multiple frames of cardiovascular angiography using the vascular skeleton. The degree of stenosis relief was determined by the change in the initial abnormal index of the target vessel segment before and after contrast agent inflow. Determine the degree of vascular improvement response of the upstream and downstream branches of the registered target vessel segment after stenosis relief. The target stenosis index of the registered target vessel segment is determined by using the degree of improvement in response of each vessel and the degree of stenosis relief, and the registered target vessel segment is classified and warned based on the target stenosis index.

[0005] Furthermore, the step of determining the vascular segment corresponding to each branch using the branch points after vascular skeletonization in cardiovascular angiography includes: Determine each skeleton node after the blood vessel skeletonization of the cardiovascular angiography image, and determine the cosine similarity of the angle between the anterior and posterior blood vessel direction vectors of the skeleton node. By comparing the cosine similarity of the included angle with a preset similarity threshold, the actual branch points in the skeleton node are determined, and the corresponding blood vessel segments are obtained by using the branch points as the boundaries of each branch.

[0006] Furthermore, the method of determining the stenotic region and its degree of stenosis by utilizing the deviation in the inner diameter of the target vascular segment, and determining the initial abnormality index of the target vascular segment using each degree of stenosis, includes: Determine the inner diameter of the blood vessel at the target sampling point in the target blood vessel segment, the average inner diameter of the window in which it is located, and the deviation of the inner diameter of the blood vessel from the average inner diameter. Compare the deviation of the blood vessel diameter with a preset deviation threshold, and use the target sampling point that is greater than the preset deviation threshold as the stenosis sampling point; Starting from the narrow sampling point, traverse along the blood vessel axis towards both ends until the deviation of the inner diameter of the blood vessels at both ends is less than or equal to the preset deviation threshold, and stop traversing to obtain the narrow region and its degree of narrowing. Determine the weighted difference ratio between the stenosis length of the narrowing region and the total vessel length of the target vessel segment, and use the weighted difference ratio and the degree of stenosis to determine the initial abnormality index of the target vessel segment.

[0007] Furthermore, obtaining the narrow region and its degree of narrowness includes: The minimum vascular diameter in the narrow region is taken as the target vascular diameter of the narrow region, and the mean normal vascular diameter of the target vascular segment is determined. The degree of narrowing in the narrow area is calculated using the average of the target pipe diameter and the normal pipe diameter.

[0008] Furthermore, the step of using the vascular skeleton to perform feature registration on multiple frames of cardiovascular angiography images to obtain the registered target vascular segment includes: The cardiovascular angiography image when the contrast agent has not filled the blood vessels is used as the baseline frame, and the cardiovascular angiography image when the contrast agent has filled the blood vessels is used as the subsequent frame. Multiple matching feature point pairs are extracted from the vascular skeleton of the reference frame and subsequent frames. The spatial distance error of the feature point pairs is minimized to obtain the registered target vascular segment after rigid registration.

[0009] Furthermore, the process of minimizing the spatial distance error of feature point pairs to obtain the rigidly registered target blood vessel segment further includes: Sample the vascular skeleton along the reference frame and subsequent frames at preset distances to obtain each skeleton sampling point, and determine the coordinate difference of each skeleton sampling point between the subsequent frame and the reference frame. The global deformation field is obtained by solving the coordinate difference using the thin plate spline difference function, and the pixel position of subsequent frames is corrected using the global deformation field to obtain the registered target blood vessel segment after elastic registration.

[0010] Furthermore, the determination of the degree of stenosis relief using the initial abnormal index changes of the registered target vessel segment before and after contrast agent inflow includes: Determine the initial abnormality index of the target vessel segment before contrast agent inflow and the minimum abnormality index after contrast agent inflow. The degree of stenosis relief in the registered target vessel segment was calculated by utilizing the difference between the initial abnormal index before contrast agent inflow and the minimum abnormal index after contrast agent inflow.

[0011] Further, determining the degree of vascular improvement response of the upstream and downstream branches of the registered target vessel segment after stenosis relief includes: The abnormality index of the upstream branch is obtained by weighting the difference between the abnormality degree of the upstream branch of the target vessel segment and the overall abnormality degree of all upstream branches, combined with the abnormality degree of the upstream branch; where the abnormality degree includes the degree of stenosis and the degree of dilation. Determine the first and second abnormality indices of the upstream branches of the target vessel segment before contrast agent inflow and after stenosis relief; The degree of vascular improvement response of the upstream branch of the registered target vascular segment is calculated using the first and second abnormality indices.

[0012] Further, determining the target stenosis index of the registered target vessel segment using the respective vessel improvement response degree and the stenosis relief degree includes: Determine the target duration of the registered target vessel segment during the stenosis relief process as a percentage of the maximum duration of relief across all vessel segments. The target stenosis index of the registered target vessel segment is calculated by using the target duration percentage, the degree of stenosis relief, and the degree of vascular improvement response of the upstream and downstream branches.

[0013] The present invention also provides an intelligent classification system for cardiovascular angiography images, the system comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the method as described in any of the above.

[0014] The present invention has the following beneficial effects: This invention divides the entire vascular network into several functional vascular segments. When classifying blood vessels using segments as the basic unit, it's difficult to comprehensively reflect the true state of blood flow within a segment if the vascular system is a continuous circulatory system and the stenosis of a single vessel is analyzed in isolation. Therefore, this invention not only extracts the degree of stenosis from the segment itself but also analyzes the features of its upper and lower branches, achieving a more precise quantitative representation of vascular stenosis and a refined classification of stenosis states. This provides more comprehensive data for the classification of vascular segments, thereby offering more scientific data references for the diagnosis and treatment of vascular diseases. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 A flowchart illustrating the steps of an intelligent classification method for cardiovascular angiography images provided in one embodiment of the present invention; Figure 2 This is a detailed flowchart of step S1 in a cardiovascular angiography image intelligent classification method provided in an embodiment of the present invention; Figure 3 This is a detailed flowchart of step S2 in a cardiovascular angiography image intelligent classification method provided in an embodiment of the present invention; Figure 4 This is a detailed flowchart of step S3 in a cardiovascular angiography image intelligent classification method provided in an embodiment of the present invention; Figure 5 This is a detailed flowchart of step S4 in a cardiovascular angiography image intelligent classification method provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the hardware operating environment of the intelligent classification device for cardiovascular angiography images involved in the embodiments of the present invention; Figure 7 This is a schematic diagram of the framework structure of the intelligent classification system for cardiovascular angiography images involved in the embodiments of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a cardiovascular angiography image intelligent classification method proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] Before proceeding with the following embodiments, it is necessary to explain the main objective of this invention: An intelligent classification system for angiographic images based on the degree of stenosis in vascular segments. After dividing the angiographic vessels into numerous segments, the system uses the changes in the characteristics of each segment and its superior and inferior branches in time-series images to achieve fine segmentation of the vascular network in the images.

[0020] The specific scenarios addressed by this invention can be: When doctors make clinical diagnoses of patients based on angiography images, the classification results generated based on the degree of vascular stenosis can help doctors quickly locate the segment of blood vessel with a high degree of stenosis, improving the efficiency and accuracy of the diagnostic process.

[0021] The specific scheme of the intelligent classification method for cardiovascular angiography images provided by the present invention will be described in detail below with reference to the accompanying drawings.

[0022] Example 1: For the intelligent classification method for cardiovascular angiography images provided by this invention, please refer to [link to relevant documentation]. Figure 1 The diagram illustrates a flowchart of the intelligent classification method for cardiovascular angiography images provided in an embodiment of the present invention.

[0023] The intelligent classification method for cardiovascular angiography images includes: Step S1: Use the branch points of the vascular skeletonized image in the cardiovascular angiography to determine the vascular segment corresponding to each branch. In this embodiment, an angiography catheter is inserted into the target cardiovascular region via a peripheral blood vessel using interventional techniques. Simultaneously with the injection of contrast agent, a CT (Computed Tomography) angiography system is used to dynamically capture the filling process of the contrast agent within the vessel, acquiring multiple consecutive high-resolution images of the vascular morphology, i.e., cardiovascular angiography. Throughout the process, the injection rate and dosage of the contrast agent are precisely controlled, and the imaging parameters are adjusted synchronously to ensure that the images clearly show details such as the vessel's course, diameter changes, and stenosis, providing a high-quality raw data foundation for subsequent vascular segment analysis and classification.

[0024] To facilitate vascular classification, the complex network structure of blood vessels needs to be divided. Therefore, by extracting the vascular centerline through vascular skeletonization and then dividing the vascular segments using skeleton nodes as boundaries, we can not only decompose the vascular network into units, but also establish a standardized spatial reference for subsequent assessment of stenosis. That is, each vascular segment is used as an independent analysis unit, and the degree of vascular stenosis is quantified by combining the characteristics of upstream and downstream branches.

[0025] Specifically, please refer to Figure 2 Step S1 includes: Step S11: Determine each skeleton node after the cardiovascular angiography image is skeletonized, and determine the cosine similarity of the angle between the anterior and posterior vessel direction vectors of the skeleton nodes. Step S12: Compare the cosine similarity of the included angle with the preset similarity threshold to determine the actual branch points in the skeleton node and obtain the corresponding blood vessel segments by using the branch points as the boundaries of each branch.

[0026] In this embodiment, when blood vessels with different orientations intersect in space, their projections will appear as branch-like shapes on the image. However, these "pseudo" branch points do not possess the structural features of real branch points. Therefore, these "pseudo" branch points need to be removed from the skeleton nodes before formal segmentation. Since the orientation of blood vessels on both sides of the intersection point is continuous in space (e.g., the main trend of blood vessels before the intersection point is left-right, and the branches after the node still maintain this trend), the direction vector v1 of the blood vessels before the intersection point and the direction vector v2 of the blood vessels after the intersection point are calculated respectively, and a discrimination formula based on the cosine similarity of the angle is constructed (existing cosine similarity formula): .when If it is a pseudo-node caused by blood vessel intersection, it is excluded; if If it is, then it is confirmed as a real branch node and used as the boundary point for dividing the blood vessel segment. The preset similarity threshold can be set to 0.7~0.9, and can be adjusted according to the actual situation.

[0027] After filtering out pseudo-nodes, the skeleton nodes all correspond to the actual branches in the vascular network. Using these branch nodes as segmentation points, the vascular network is segmented to obtain the corresponding vascular segments, which can provide a structural basis for subsequent vascular classification based on the degree of stenosis.

[0028] Step S2: Determine the stenotic region and its degree of stenosis by using the deviation of the inner diameter of the target blood vessel segment, and determine the initial abnormality index of the target blood vessel segment by using each degree of stenosis. Before quantifying the stenosis of a target vessel segment (any vessel segment), it is necessary to first calculate the diameter of the normal area within that segment (i.e., the normal diameter baseline D0). This is because the normal diameter is the core reference for judging stenosis within a vessel segment. Only by clearly defining the normal diameter of a vessel under physiological conditions can the severity of stenosis be quantified by the "degree of deviation between the actual diameter and the normal diameter." Without this baseline, the determination of stenosis will lack a unified reference, making it impossible to accurately distinguish between normal physiological fluctuations and stenotic areas.

[0029] Specifically, please refer to Figure 3 Step S2 includes: Step S21: Determine the inner diameter of the blood vessel at the target sampling point in the target blood vessel segment, the average inner diameter of the window in which it is located, and the deviation of the inner diameter of the blood vessel from the average inner diameter. Step S22: Compare the deviation of the blood vessel diameter with a preset deviation threshold, and take the target sampling point that is greater than the preset deviation threshold as the stenosis sampling point. Step S23: Starting from the narrow sampling point, traverse along the blood vessel axis to both ends until the deviation of the inner diameter of the blood vessels at both ends is less than or equal to the preset deviation threshold, and stop traversing to obtain the narrow region and its degree of narrowing. Specifically, step S23, which obtains the narrow region and its degree of narrowness, includes: The minimum vascular diameter in the narrow region is taken as the target vascular diameter of the narrow region, and the mean normal vascular diameter of the target vascular segment is determined. The degree of narrowing in the narrow area is calculated using the average of the target pipe diameter and the normal pipe diameter.

[0030] Step S24: Determine the difference ratio weight between the stenosis length of the stenosis region and the total vascular length of the target vascular segment, and use the difference ratio weight and the degree of stenosis to determine the initial abnormality index of the target vascular segment.

[0031] In this embodiment, sampling points are evenly distributed within the blood vessel segment (the number of sampling points can be selected as 10-20, and can be adjusted according to the actual situation). At each sampling point P... h At this location, the inner diameter of the blood vessel perpendicular to the vessel axis is measured and recorded as follows. Because blood vessels are tubular structures, their diameter does not change drastically under normal physiological conditions. Therefore, abnormal points that deviate from the local trend can be identified through local smoothing.

[0032] Choose a window width l, for example, l=3 means that the local reference value of each sampling point is calculated from itself and one point before and after it. For the h-th sampling point (as the target sampling point, referring to any sampling point), the local smoothing value (the average inner diameter of the window) is... (Edge points can reduce window size, such as the first point using...) ), calculate the absolute deviation (vessel diameter deviation) between the actual diameter and the local smoothed value at all sampling points. The 90th quantile of the deviation value is taken as the threshold E for local fluctuation (a preset deviation threshold, which can be adjusted), that is, the deviation of 90% of normal (sampling) points is ≤ E. For any sampling point, if If the diameter is ≤E, it is considered a normal pipe diameter, and the corresponding target sampling point is a normal sampling point; otherwise, it is a narrow pipe diameter and a narrow sampling point.

[0033] For the sampling point P that is determined to be narrow h From P h Starting from a point, traverse along the vessel axis (obtained through skeletal mapping) towards both ends, recording the vessel diameter along the path. Stop traversing when both ends of the traversal have normal vessel diameters, thus identifying the corresponding narrowed region. Record the smallest diameter D within this narrowed region during the traversal. min The minimum vessel diameter is used as the representative diameter at the stenosis, which is also the target diameter. A spatial weighting factor (difference ratio weight) is also introduced. For the i-th narrow region, the length of the traversed narrow region is denoted as the narrow length L. i The total length of the vessel segment is denoted as the total vessel length L. A The weighting formula is: For abnormal (stenosis) conditions of vascular segment stenosis The value that can be assigned is: In the formula, This represents the narrowest pipe diameter in narrow region i, and can be used as the degree of narrowness of narrow region i: ,in This is the average value of all pipe diameters that meet the normal pipe diameter requirements (normal pipe diameter mean). This represents the minimum internal diameter of the narrowed area, i.e., the target diameter. Narrowing of a vascular segment is a pathological abnormality and is assigned a negative value here. In cases where multiple stenotic regions may exist within a vascular segment, the vessel values ​​for these multiple stenotic regions can be obtained by superposition: Here, 'n' represents the total number of narrowed regions in the target blood vessel segment. As the initial stenosis index of the target vascular segment, it belongs to the initial abnormality index. The initial abnormality index can be divided into the initial stenosis index and the initial dilation index. In fact, the abnormality in each embodiment refers to stenosis or dilation.

[0034] Step S3: Use the vascular skeleton to perform feature registration on multiple frames of cardiovascular angiography images to obtain the registered target vascular segment, and use the change in the initial abnormal index of the registered target vascular segment before and after contrast agent inflow to determine its stenosis relief degree. In the above embodiments, an abnormality index of each vascular segment extracted from a single frame image was used to construct a quantitative basis for vascular structure. Based on this, multi-frame tracking analysis was performed on these indicators to further capture the evolution of blood vessels over time: when contrast agent flows in, the blood flow impact may cause a temporary physiological expansion of the originally narrowed area, temporarily alleviating the narrowing, and the corresponding abnormality index decreases accordingly; if the narrowing of a vascular segment is temporarily relieved by the blood flow impact, the expansion state of upstream branches caused by blood flow congestion will also be reduced accordingly, and the abnormality index of expansion will decrease accordingly; when the narrowing of a vascular segment improves due to blood flow, the blood flow of downstream branches increases, and the small narrowings originally caused by ischemia will be alleviated, and the abnormality index of narrowing will decrease accordingly. However, before capturing these dynamic changes, it is necessary to exclude interference from physiological activities such as heartbeat, as these periodic vibrations can cause non-pathological fluctuations in vascular images, which may mask the true blood flow changes when contrast agent passes through.

[0035] Heartbeats can cause overall displacement or localized deformation of blood vessels. This movement manifests as inconsistent spatial positions and morphologies of the same vascular structure across multiple frames. For example, a narrowed area of ​​a blood vessel segment in one frame might be misjudged as a normally shifted area in another frame due to displacement caused by heartbeats, or misjudged as a change in the degree of narrowing due to deformation. To eliminate spatial positional deviations of vascular structures caused by physiological movements (such as heartbeats) across multiple frames, it is necessary to calibrate these deviations to ensure spatial alignment of the vascular structures in the images.

[0036] Specifically, please refer to Figure 4 Step S3, which involves using the vascular skeleton to perform feature registration on multiple frames of cardiovascular angiography images to obtain the registered target vascular segment, includes: Step S31: Use the cardiovascular angiography image when the contrast agent has not filled the blood vessels as the reference frame, and use the subsequent cardiovascular angiography image when the contrast agent has filled the blood vessels as the subsequent frame. Step S32: Extract multiple matching feature point pairs from the vascular skeleton of the reference frame and subsequent frames, and minimize the spatial distance error of the feature point pairs to obtain the registered target vascular segment after rigid registration.

[0037] In this embodiment, a cardiovascular angiography image before the contrast agent fills the vessel is selected as the reference frame where the vessel morphology is relatively stable. At this time, the vessel segment has not received dynamic images of the contrast agent flow, and is close to the natural static physiological state of the vessel, which can be used as a spatial template for subsequent frames. Correspondingly, a cardiovascular angiography image after the contrast agent fills the vessel is used as the subsequent frame.

[0038] Because the vascular skeleton can stably represent the course and branching connections of blood vessels, its structure remains consistent even if the vessels shift or deform due to heartbeat. Therefore, the vascular skeleton (the central axis of the blood vessel) is chosen as the registration feature.

[0039] The overall displacement caused by the heartbeat is the most obvious source of error. Since the vascular structure (e.g., branch connections, skeleton orientation) remains unchanged across different frames, only its spatial location differs. Rigid registration can be used to align the reference frame with subsequent frames through geometric transformations: extract q matching feature point pairs from the vascular skeleton of the reference frame and subsequent frames. The feature point set of the reference frame is represented as... The feature point set of subsequent frames is represented as The goal is to minimize the spatial distance error between feature point pairs, and the vessel displacement registration error function is defined as follows: In the formula, The function representing the vessel displacement registration error is shown in parentheses. These represent the rotation angle, x-axis translation, and y-axis translation that need to be optimized, respectively; here, q represents the number of feature point pairs. This represents the coordinates of the k-th feature point in the reference frame; This represents the coordinates of the k-th feature point in a subsequent frame after rotation and translation transformation; This represents the square of the Euclidean distance between two coordinates (any pair of feature points). The least squares method is used to find the minimum value of E. The blood vessel pixels in subsequent frames are then registered as a whole to obtain the rigidly registered target blood vessel segment.

[0040] After step S32, the method further includes: Sample the vascular skeleton along the reference frame and subsequent frames at preset distances to obtain each skeleton sampling point, and determine the coordinate difference of each skeleton sampling point between the subsequent frame and the reference frame. The global deformation field is obtained by solving the coordinate difference using the thin plate spline difference function, and the pixel position of subsequent frames is corrected using the global deformation field to obtain the registered target blood vessel segment after elastic registration.

[0041] In this embodiment, the rigid registration mentioned in the previous embodiment cannot handle local deformations caused by heartbeats (such as the periodic expansion and contraction of the vessel wall with blood pressure fluctuations, and the slight bending of branch vessels). These deformations will cause the vascular morphology of the same vessel segment to differ between the reference frame and subsequent frames, requiring further correction through elastic registration. In the subsequent frames and the reference frame after rigid registration, sampling is performed along the vascular skeleton at preset fixed intervals (e.g., 0.5 mm, which can be adjusted). The coordinate difference between each skeleton sampling point (referred to as the skeleton sampling point here for easy distinction from the above sampling points) in the subsequent frames and the reference frame is used to solve the global deformation field through the TPS (Thin Plate Spline) difference function. The global deformation field is applied to each blood vessel pixel in subsequent frames, and the corrected pixel position is obtained through resampling. This process ultimately ensures that the local morphology of the blood vessels in subsequent frames is consistent with that of the reference frame, resulting in the elastically registered target blood vessel segment.

[0042] Furthermore, step S3, which determines the degree of stenosis relief by utilizing the initial abnormal index change of the registered target vessel segment before and after contrast agent inflow, specifically includes: Determine the initial abnormality index of the target vessel segment before contrast agent inflow and the minimum abnormality index after contrast agent inflow. The degree of stenosis relief in the registered target vessel segment was calculated by utilizing the difference between the initial abnormal index before contrast agent inflow and the minimum abnormal index after contrast agent inflow.

[0043] For the registered target vessel segment itself, after the contrast agent flows in, the stenosis index of the vessel segment decreases from its original value to its minimum value. The relative amplitude during this process reflects the dilating effect of blood flow on the stenotic area (called the degree of stenosis relief). ,in The initial stenosis index (initial abnormality index) is the vascular segment before contrast agent inflow. It is the minimum value of the narrowing index during or after the contrast agent inflow process, i.e., the minimum abnormality index. This indicates the difference in change between the two. Degree of stenosis relief. A larger value indicates a more significant relief of stenosis. It should be noted that, to prevent this from being used as the denominator... For cases where the denominator is zero, a minimum value 'a' can be added, for example, a = 0.000001. Furthermore, this minimum value 'a' can be set for the denominators in all formulas within the context to avoid the extreme case of a denominator being zero.

[0044] Step S4: Determine the degree of vascular improvement response of the upstream and downstream branches of the registered target vessel segment after stenosis relief. When classifying vascular segments in cardiovascular angiography, the completeness of feature representation is crucial. Since the vascular system is an interconnected circulatory whole, extracting features from only a single segment may result in incomplete feature acquisition. Therefore, when extracting features from a single segment, in addition to focusing on the segment's own characteristics, it is also necessary to consider the structural features of its upstream and downstream branches to achieve a comprehensive representation of the vascular functional state.

[0045] In angiography, the upstream branches of a vessel segment are the starting points for blood flow. The state of these branches is directly related to the degree of stenosis in the downstream segment. When stenosis exists in the downstream segment, the upstream branches may dilate due to blood flow congestion. The degree of stenosis within the vessel segment itself can be directly reflected by the difference between the stenotic area and the normal area. The state of the downstream branches is also affected by the stenosis of the upstream segment, and may exhibit mild stenosis due to reduced blood perfusion. These three types of features are interconnected: the dilation of the upstream branches is a compensatory response to the stenosis of the downstream segment; the stenosis of the vessel segment itself directly determines the degree of influence; and the stenosis of the downstream branches is a chain reaction caused by the stenosis of the upstream. Extracting features from these three dimensions can more comprehensively represent the stenosis state of the vessel.

[0046] Specifically, please refer to Figure 5 Step S4 includes: Step S41: Using the weighted difference ratio between the abnormality degree of the upstream branch of the target vessel segment and the overall abnormality degree of all upstream branches, the abnormality index of the upstream branch is obtained by combining the abnormality degree of the upstream branch; wherein, the abnormality degree includes the degree of stenosis and the degree of dilation. Step S42: Determine the first and second abnormal indices of the upstream branches of the registered target vessel segment before contrast agent inflow and after stenosis relief. Step S43: Using the first abnormality index and the second abnormality index, calculate the degree of vascular improvement response of the upstream branch of the registered target vascular segment.

[0047] In this embodiment, stenosis of a vascular segment has different effects on its upstream and downstream branches. Upstream branches (branches where blood enters the segment) will dilate due to obstructed blood flow caused by the stenosis; downstream branches (branches where blood leaves the segment) will contract slightly due to reduced blood supply caused by the stenosis. Therefore, this effect is quantified by considering the number and primary / secondary differences of the branches. The degree of dilation or contraction at a branch is directly related to the vascular segment in which the branch is located (for example, a branch at the upper branch essentially belongs to the lower half of the previous segment). Furthermore, since there is more than one branch, the branches need to be superimposed based on their diameter differences to accurately quantify the degree of abnormality at the branch (positive for dilation, negative for stenosis).

[0048] For each branch (regardless of its upper or lower position), calculate the ratio of the diameter difference between the u-th branch vessel and the normal branch vessels in its parent main vessel (or segment) (i.e., the degree of abnormality, analogous to the degree of stenosis in the target segment mentioned above; the degree of abnormality is also divided into stenosis or dilation): Degree of Abnormality ,in This represents the average diameter of normal branch vessels within the vascular segment to which the u-th branch vessel belongs. This represents the diameter of the abnormal region. The method for determining the abnormal region is the same as for the stenosis region mentioned above. The corresponding diameter can be the average of the vessel diameters at multiple sampling points within the abnormal region, or it can be the smallest vessel diameter. Since there is usually more than one branch vessel at a branch point, it is necessary to quantify the degree of abnormality (dilation or constriction) at the branch point by combining all branch vessels. The greater the difference between the branch diameter and the normal branch diameter in the main vessel, the greater the contribution of that branch to the overall abnormality. Therefore, a difference ratio weighting is designed. Here, m represents the number of branch vessels at the branch point. This represents the overall anomaly level of all upstream or downstream branches at a given point (a minimum value 'a' can also be set to prevent the denominator from being zero when it is used as the denominator), and the anomaly index of the corresponding upstream or downstream branch at that point. for: In the formula, Indicates the degree of abnormality in the upstream or downstream branch. This indicates the weight of the difference between upstream and downstream branches; Indicates the direction factor. For the upstream branches (the main trend is expansion). The value is +1; for downstream branches (the main trend is contraction). The value is -1. It needs to be explained again that here... Uniformity refers to the vascular diameter at the sampling point, regardless of the target vascular segment or its upstream and downstream branches. This refers to the average normal diameter of each blood vessel segment (the upstream and downstream branches correspond to the normal branch blood vessel diameters of their respective segments).

[0049] By extracting features from upstream branches, the vessel segment itself, and downstream branches, an assessment system covering the entire vessel segment is formed. These three aspects corroborate each other, enabling a quantitative assessment of vascular health status in a single frame image. Further multi-frame comparative analysis can capture the dynamic changes in vascular structure over time, providing a more scientific basis for clinical decision-making.

[0050] Furthermore, the dynamic characteristics at the upper branch are quantified. After the stenosis of the vascular segment is relieved, the decrease in the dilation index of the upstream branch due to the reduction in blood flow congestion reflects the degree of response of the upstream to the improvement in downstream blood flow, that is, the degree of vascular improvement response of the upstream. ,in The dilation index of the upstream branch anterior to the target vessel segment into which contrast agent flows (registered) is denoted here as the first abnormality index. The dilation index of the upstream branch after the stenosis of the target vessel segment is relieved is denoted here as the second abnormal index. The larger the value, the more significant the improvement in the expansion state of the upstream branch.

[0051] Similarly, the dynamic characteristics at the lower branches are quantified. After the stenosis in the vascular segment is relieved, the decrease in the stenosis index in the downstream branches due to increased blood flow reflects the transmission effect of the improved blood flow: the degree of improvement in the downstream vascular response. ,in This is an index of stenosis in the downstream branches anterior to the target vessel segment where contrast agent can flow. It can be denoted as the third abnormality index. After the stenosis of the target vessel segment is relieved, the stenosis index of the downstream branches can be denoted as the fourth abnormality index. The higher the value, the more significant the improvement in the ischemic state of the downstream branches.

[0052] In addition, to prevent the denominator from being used as a denominator and In the extreme case where the value is 0, a very small value 'a' can be added, for example, a = 0.000001.

[0053] Step S5: Determine the target stenosis index of the registered target vessel segment by using the respective vessel improvement response degree and the stenosis relief degree, and classify and warn the registered target vessel segment based on the target stenosis index.

[0054] Specifically, step S5, which determines the target stenosis index of the registered target vessel segment using the respective vessel improvement response degree and the stenosis relief degree, includes: Determine the target duration of the registered target vessel segment during the stenosis relief process as a percentage of the maximum duration of relief across all vessel segments. The target stenosis index of the registered target vessel segment is calculated by using the target duration percentage, the degree of stenosis relief, and the degree of vascular improvement response of the upstream and downstream branches.

[0055] In this embodiment, the length of time the vascular segment stenosis index remains in a relieved state (target relief duration) is defined as the duration of time the stenosis index remains in a relieved state. This can reflect the duration of the blood flow impact effect: In the formula, This indicates the time point at which the stenosis index of the target vessel segment recovers to more than 90% of its original value (after registration). This indicates the point in time when its narrowing index reaches its minimum value. The higher the value, the more sustained the blood flow's dilating effect on the narrowed area.

[0056] By fusing the change rates of three characteristics—the target vascular segment, its upstream and downstream branches—a comprehensive evaluation index can be constructed, which can more completely depict the changes in vascular status over time. Considering the complexity of vascular network structures, some vascular segments may lack upstream branches (such as the origin of the aorta) or downstream branches (such as terminal arterioles). Therefore, a formula is designed to address the situation of missing branches and calculate the target stenosis index. : In the formula, This indicates the baseline stenosis index calculated based on the target vessel segment itself: .in, This represents the maximum duration of remission across all vascular segments. It is used for normalization and is generally not zero. If there is an extreme case where it is zero, the aforementioned minimum value 'a' can be added to prevent the denominator from being zero. Indicates the percentage of target duration. Degree of stenosis relief. The larger the value, the more significant the relief of narrowing. Positively correlated with the degree of narrowing. Target relief duration. The larger the value, the longer the relief lasts, and the more likely the narrowing is to be temporary. It is positively correlated with the degree of narrowness.

[0057] The correction term indicating the degree of improvement in response of upstream branch vessels: ; The correction term indicating the degree of improvement in response of downstream branch vessels: ; If branches exist or The more significant the improvement in branching (the greater the degree of vascular improvement response) The larger the value, the smaller the correction term, and the lower the final narrowing index.

[0058] Calculate (register) the target stenosis index of the target vessel segment By integrating the vascular changes of the vessel segment itself and its superior and inferior branches during the contrast agent flow process, it is possible to comprehensively capture the dynamic stability of the vascular system from blood flow input to output, providing a quantitative tool that integrates multi-dimensional information for vascular health assessment.

[0059] Based on the final target narrowing index Classify the blood vessel segments. Divide each blood vessel segment into... A cumulative distribution curve was plotted, and the vessel segments were divided into three categories based on the cumulative distribution curve. The 75th percentile was used as the threshold for distinguishing between medium and high risk. That is, 75% of the samples in the cumulative distribution. Values ​​less than this value; the 25th percentile is used as the low-to-medium risk threshold. That is, 25% of the samples in the cumulative distribution. The value is less than this value.

[0060] At that time, the corresponding blood vessel segment will be marked as a high-risk stenosis, requiring close monitoring by the doctor; At that time, it was marked as a medium-risk area and required regular monitoring; It is marked as low-risk and narrow, requiring no special attention.

[0061] This invention divides the entire vascular network into several functional vascular segments. When classifying blood vessels using segments as the basic unit, it's difficult to comprehensively reflect the true state of blood flow within a segment if the vascular system is a continuous circulatory system and the stenosis of a single vessel is analyzed in isolation. Therefore, this invention not only extracts the degree of stenosis from the segment itself but also analyzes the features of its upper and lower branches, achieving a more precise quantitative representation of vascular stenosis and a refined classification of stenosis states. This provides more comprehensive data for the classification of vascular segments, thereby offering more scientific data references for the diagnosis and treatment of vascular diseases.

[0062] Example 2: This invention also proposes an intelligent classification device for cardiovascular angiography images. The device can be a computer, a server, or a combination of multiple devices for data analysis and computation.

[0063] like Figure 6 As shown, Figure 6 This is a schematic diagram of the hardware operating environment of the intelligent classification device for cardiovascular angiography images involved in the embodiments of the present invention.

[0064] like Figure 6As shown, the intelligent classification device for cardiovascular angiography images may include: a processor 1001, such as a CPU; a network interface 1004; a user interface 1003; a memory 1005; and a communication bus 1002. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display or an input unit such as a control panel; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001. The memory 1005, as a computer storage medium, may include an intelligent classification program for cardiovascular angiography images.

[0065] Those skilled in the art will understand that Figure 6 The hardware structure shown does not constitute a limitation on the device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0066] Continue to refer to Figure 6 , Figure 6 The memory 1005, which is a computer-readable storage medium, may include an operating system, a user interface module, a network communication module, and a cardiovascular angiography image intelligent classification program.

[0067] exist Figure 6 In this embodiment, the network communication module is mainly used to connect to the server and can communicate with the server for data; while the processor 1001 can call the cardiovascular angiography image intelligent classification program stored in the memory 1005 and execute the steps in the above embodiments.

[0068] Based on the hardware structure of the above-mentioned intelligent classification device for cardiovascular angiography images, various embodiments of the intelligent classification method for cardiovascular angiography images of the present invention are implemented.

[0069] In addition, the present invention also provides an intelligent classification system for cardiovascular angiography images, please refer to... Figure 7 The cardiovascular angiography image intelligent classification system includes: The vascular region segmentation module A10 is used to determine the vascular segment corresponding to each branch by using the branch points of the vascular skeleton after cardiovascular angiography; to determine the stenosis area and its degree of stenosis by using the vascular diameter deviation in the target vascular segment; and to determine the initial abnormality index of the target vascular segment by using each degree of stenosis. The vascular change analysis module A20 is used to perform feature registration on multiple frames of cardiovascular angiography images using the vascular skeleton to obtain the registered target vascular segment, and to determine the degree of stenosis relief by using the initial abnormal index change of the registered target vascular segment before and after contrast agent inflow; and to determine the degree of vascular improvement response of the upstream and downstream branches of the registered target vascular segment after stenosis relief. The vascular classification and early warning module A30 is used to determine the target stenosis index of the registered target vascular segment by using the respective vascular improvement response degree and the stenosis relief degree, and to classify and warn the registered target vascular segment based on the target stenosis index.

[0070] Furthermore, the blood vessel region segmentation module A10 is also used for: Determine each skeleton node after the blood vessel skeletonization of the cardiovascular angiography image, and determine the cosine similarity of the angle between the anterior and posterior blood vessel direction vectors of the skeleton node. By comparing the cosine similarity of the included angle with a preset similarity threshold, the actual branch points in the skeleton node are determined, and the corresponding blood vessel segments are obtained by using the branch points as the boundaries of each branch.

[0071] Furthermore, the blood vessel region segmentation module A10 is also used for: Determine the inner diameter of the blood vessel at the target sampling point in the target blood vessel segment, the average inner diameter of the window in which it is located, and the deviation of the inner diameter of the blood vessel from the average inner diameter. Compare the deviation of the blood vessel diameter with a preset deviation threshold, and use the target sampling point that is greater than the preset deviation threshold as the stenosis sampling point; Starting from the narrow sampling point, traverse along the blood vessel axis towards both ends until the deviation of the inner diameter of the blood vessels at both ends is less than or equal to the preset deviation threshold, and stop traversing to obtain the narrow region and its degree of narrowing. Determine the weighted difference ratio between the stenosis length of the narrowing region and the total vessel length of the target vessel segment, and use the weighted difference ratio and the degree of stenosis to determine the initial abnormality index of the target vessel segment.

[0072] Furthermore, the blood vessel region segmentation module A10 is also used for: The minimum vascular diameter in the narrow region is taken as the target vascular diameter of the narrow region, and the mean normal vascular diameter of the target vascular segment is determined. The degree of narrowing in the narrow area is calculated using the average of the target pipe diameter and the normal pipe diameter.

[0073] Furthermore, the vascular change analysis module A20 is also used for: The cardiovascular angiography image when the contrast agent has not filled the blood vessels is used as the baseline frame, and the cardiovascular angiography image when the contrast agent has filled the blood vessels is used as the subsequent frame. Multiple matching feature point pairs are extracted from the vascular skeleton of the reference frame and subsequent frames. The spatial distance error of the feature point pairs is minimized to obtain the registered target vascular segment after rigid registration.

[0074] Furthermore, the vascular change analysis module A20 is also used for: Sample the vascular skeleton along the reference frame and subsequent frames at preset distances to obtain each skeleton sampling point, and determine the coordinate difference of each skeleton sampling point between the subsequent frame and the reference frame. The global deformation field is obtained by solving the coordinate difference using the thin plate spline difference function, and the pixel position of subsequent frames is corrected using the global deformation field to obtain the registered target blood vessel segment after elastic registration.

[0075] Furthermore, the vascular change analysis module A20 is also used for: Determine the initial abnormality index of the target vessel segment before contrast agent inflow and the minimum abnormality index after contrast agent inflow. The degree of stenosis relief in the registered target vessel segment was calculated by utilizing the difference between the initial abnormal index before contrast agent inflow and the minimum abnormal index after contrast agent inflow.

[0076] Furthermore, the vascular change analysis module A20 is also used for: The abnormality index of the upstream branch is obtained by weighting the difference between the abnormality degree of the upstream branch of the target vessel segment and the overall abnormality degree of all upstream branches, combined with the abnormality degree of the upstream branch; where the abnormality degree includes the degree of stenosis and the degree of dilation. Determine the first and second abnormality indices of the upstream branches of the target vessel segment before contrast agent inflow and after stenosis relief; The degree of vascular improvement response of the upstream branch of the registered target vascular segment is calculated using the first and second abnormality indices.

[0077] Furthermore, the blood vessel classification and early warning module A30 is also used for: Determine the target duration of the registered target vessel segment during the stenosis relief process as a percentage of the maximum duration of relief across all vessel segments. The target stenosis index of the registered target vessel segment is calculated by using the target duration percentage, the degree of stenosis relief, and the degree of vascular improvement response of the upstream and downstream branches.

[0078] The specific implementation of the intelligent classification system for cardiovascular angiography images of the present invention is basically the same as the embodiments of the intelligent classification method for cardiovascular angiography images described above, and will not be repeated here.

[0079] Furthermore, the present invention also provides a computer-readable storage medium. The computer-readable storage medium stores a cardiovascular angiography image intelligent classification program, wherein, when executed by a processor, the cardiovascular angiography image intelligent classification program implements the steps of the cardiovascular angiography image intelligent classification method as described above.

[0080] The method implemented when the intelligent classification program for cardiovascular angiography images is executed can be referred to in various embodiments of the intelligent classification method for cardiovascular angiography images of the present invention, and will not be repeated here.

[0081] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0082] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0083] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0084] The above description is only a preferred embodiment of the present invention and does not limit the scope of protection of the present invention. All equivalent structural / method transformations made under the inventive concept of the present invention using the contents of the present invention specification and drawings, or direct / indirect applications in other related technical fields, are included within the scope of protection of the present invention.

Claims

1. A method for intelligent classification of cardiovascular angiography images, characterized in that, The method includes the following steps: The vascular segment corresponding to each branch is determined by the branch points of the vascular skeletonized image in the cardiovascular angiography. The stenotic region and its degree of stenosis are determined by the deviation of the inner diameter of the target vascular segment, and the initial abnormality index of the target vascular segment is determined by each degree of stenosis. The target vessel segment was obtained by feature registration of multiple frames of cardiovascular angiography using the vascular skeleton. The degree of stenosis relief was determined by the change in the initial abnormal index of the target vessel segment before and after contrast agent inflow. Determine the degree of vascular improvement response of the upstream and downstream branches of the registered target vessel segment after stenosis relief; The target stenosis index of the registered target vessel segment is determined by using the degree of improvement in response of each vessel and the degree of stenosis relief, and the registered target vessel segment is classified and warned based on the target stenosis index.

2. The intelligent classification method for cardiovascular angiography images according to claim 1, characterized in that, The method of determining the vascular segment corresponding to each branch using the branch points after vascular skeletonization in cardiovascular angiography includes: Determine each skeleton node after the blood vessel skeletonization of the cardiovascular angiography image, and determine the cosine similarity of the angle between the anterior and posterior blood vessel direction vectors of the skeleton node. By comparing the cosine similarity of the included angle with a preset similarity threshold, the actual branch points in the skeleton node are determined, and the corresponding blood vessel segments are obtained by using the branch points as the boundaries of each branch.

3. The intelligent classification method for cardiovascular angiography images according to claim 1, characterized in that, The method of determining the stenotic region and its degree by utilizing the deviation of the vascular diameter in the target vascular segment, and determining the initial abnormality index of the target vascular segment using each degree of stenosis, includes: Determine the inner diameter of the blood vessel at the target sampling point in the target blood vessel segment, the average inner diameter of the window in which it is located, and the deviation of the inner diameter of the blood vessel from the average inner diameter. Compare the deviation of the blood vessel diameter with a preset deviation threshold, and use the target sampling point that is greater than the preset deviation threshold as the stenosis sampling point; Starting from the narrow sampling point, traverse along the blood vessel axis towards both ends until the deviation of the inner diameter of the blood vessels at both ends is less than or equal to the preset deviation threshold, and stop traversing to obtain the narrow region and its degree of narrowing. Determine the weighted difference ratio between the stenosis length of the narrowing region and the total vascular length of the target segment, and use the weighted difference ratio and the degree of stenosis to determine the initial abnormality index of the target segment.

4. The intelligent classification method for cardiovascular angiography images according to claim 3, characterized in that, The acquisition of the narrow region and its degree of narrowness includes: The minimum vascular diameter in the narrow region is taken as the target vascular diameter of the narrow region, and the mean normal vascular diameter of the target vascular segment is determined. The degree of narrowing in the narrow area is calculated using the average of the target pipe diameter and the normal pipe diameter.

5. The intelligent classification method for cardiovascular angiography images according to claim 1, characterized in that, The method of using the vascular skeleton to perform feature registration on multiple frames of cardiovascular angiography images to obtain the registered target vascular segment includes: The cardiovascular angiography image when the contrast agent has not filled the blood vessels is used as the baseline frame, and the cardiovascular angiography image when the contrast agent has filled the blood vessels is used as the subsequent frame. Multiple matching feature point pairs are extracted from the vascular skeleton of the reference frame and subsequent frames. The spatial distance error of the feature point pairs is minimized to obtain the registered target vascular segment after rigid registration.

6. The intelligent classification method for cardiovascular angiography images according to claim 5, characterized in that, The process of minimizing the spatial distance error of feature point pairs to obtain the rigidly registered target blood vessel segment further includes: Sample the vascular skeleton along the reference frame and subsequent frames at preset distances to obtain each skeleton sampling point, and determine the coordinate difference of each skeleton sampling point between the subsequent frame and the reference frame. The global deformation field is obtained by solving the coordinate difference using the thin plate spline difference function, and the pixel position of subsequent frames is corrected using the global deformation field to obtain the registered target blood vessel segment after elastic registration.

7. The intelligent classification method for cardiovascular angiography images according to claim 1, characterized in that, The method of determining the degree of stenosis relief by utilizing the initial abnormal index changes of the registered target vessel segment before and after contrast agent infusion includes: Determine the initial abnormality index of the target vessel segment before contrast agent inflow and the minimum abnormality index after contrast agent inflow. The degree of stenosis relief in the registered target vessel segment was calculated by utilizing the difference between the initial abnormal index before contrast agent inflow and the minimum abnormal index after contrast agent inflow.

8. The intelligent classification method for cardiovascular angiography images according to claim 1, characterized in that, The determination of the degree of vascular improvement response of the upstream and downstream branches of the registered target vessel segment after stenosis relief includes: The abnormality index of the upstream branch is obtained by weighting the difference between the abnormality degree of the upstream branch of the target vessel segment and the overall abnormality degree of all upstream branches, combined with the abnormality degree of the upstream branch; where the abnormality degree includes the degree of stenosis and the degree of dilation. Determine the first and second abnormality indices of the upstream branches of the target vessel segment before contrast agent inflow and after stenosis relief. The degree of vascular improvement response of the upstream branch of the registered target vascular segment is calculated using the first and second abnormality indices.

9. The intelligent classification method for cardiovascular angiography images according to claim 1, characterized in that, The determination of the target stenosis index for the registered target vessel segment using the respective vessel improvement response degree and the stenosis relief degree includes: Determine the target duration of the registered target vessel segment during the stenosis relief process as a percentage of the maximum duration of relief across all vessel segments. The target stenosis index of the registered target vessel segment is calculated by using the target duration percentage, the degree of stenosis relief, and the degree of vascular improvement response of the upstream and downstream branches.

10. A cardiovascular angiography image intelligent classification system, the system comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 9.

Images