Blood vessel stenosis degree grading system based on coronary angiography image

Abstract

The invention relates to the technical field of blood vessel stenosis grading, in particular to a blood vessel stenosis degree grading system based on coronary angiography images. The method comprises the following steps: dividing a vascular skeleton line of an angiography image into vascular segments based on a bifurcation point, and determining a stenosis point on the vascular segment according to the blood flow velocity difference of the vascular segment at the front edge position of a contrast agent in each frame of angiography image and a neighborhood angiography image; according to the blood vessel diameter difference between each bifurcation point and the other bifurcation points and the distance between the other bifurcation points and the main blood vessel section, a reference adaptation degree is obtained, a reference point of each bifurcation point is selected, and according to the blood flow speed difference between each bifurcation point and the branch blood vessel section of the reference point and the reference adaptation degree, narrow points in the bifurcation points are screened; and grading the blood vessel stenosis degree based on the blood vessel diameter of the stenosis point. The blood vessel skeleton line is divided into two conditions of the blood vessel segment and the bifurcation point to analyze the blood vessel physiological stenosis position, and the accuracy of positioning the pathological stenosis position in the coronary artery is improved.

Description

Technical Field

[0001] This invention relates to the field of vascular stenosis grading technology, specifically to a vascular stenosis grading system based on coronary angiography images. Background Technology

[0002] Coronary artery disease (CAD) is one of the leading causes of cardiovascular death worldwide. A key aspect of its clinical diagnosis and treatment lies in assessing the degree of vascular stenosis on coronary angiography images. This assessment evaluates the extent of coronary artery disease, providing valuable information to clinicians and helping to develop appropriate treatment plans. Therefore, the grading of vascular stenosis is directly related to the selection of treatment options.

[0003] In existing technologies, the grading of vascular stenosis in coronary angiography images is mainly based on the proportion of narrowing in the diameter or area of ​​the vessel. It can also be assessed using the Thrombolysis In Myocardial Infarction (TIMI) grading criteria to determine the location and severity of stenosis in the coronary arteries. However, due to the complexity of vascular structure, the limitations of two-dimensional projection imaging, and image ambiguity, it is difficult to accurately determine the location of pathological stenosis in coronary arteries using coronary angiography images. Furthermore, the number and angle differences of branch vessels at bifurcation points lead to significant variations in blood flow velocity, making it difficult to determine whether pathological stenosis exists at bifurcation points based on blood flow velocity alone. This results in inaccurate determination of the location of pathological stenosis in the coronary arteries, leading to a low accuracy rate in grading the degree of vascular stenosis. Summary of the Invention

[0004] To address the technical problem of inaccurate location determination of pathological stenosis in coronary arteries due to factors such as vascular bifurcation and image blurring, the present invention aims to provide a vascular stenosis grading system based on coronary angiography images. The specific technical solution adopted is as follows: This invention proposes a grading system for the degree of vascular stenosis based on coronary angiography images, the system comprising: The data acquisition module is used to acquire coronary angiography video after the injection of contrast agent, the video consisting of angiography images; The intravascular stenosis localization module is used to extract the vascular skeleton line of the vascular region in the angiography image, divide the vascular skeleton line into vascular segments based on the bifurcation point, and determine the stenosis point on each vascular segment based on the difference in blood flow velocity at the contrast agent front position in each frame of angiography image and its neighboring angiography images. The stenosis localization module at the bifurcation point is used to acquire the main vessel segment, obtain the reference fit degree between each bifurcation point and the other bifurcation points based on the difference in vessel diameter between each bifurcation point and the other bifurcation points, and select a reference point for each bifurcation point; and filter out stenosis points among the bifurcation points based on the difference in blood flow velocity between the branch vessel segments of each bifurcation point and its reference point and the reference fit degree. The stenosis grading module is used to grade the degree of vascular stenosis based on the diameter of the stenosis point.

[0005] Furthermore, determining the stenosis point on each blood vessel segment includes: For each vessel segment, the angiographic images during the time period when the contrast agent flows through the vessel segment are recorded as the analysis images; The image acquisition frequency is obtained, and the ratio of the distance between the blood vessel segment and the position of the contrast agent leading edge in each frame of the analysis image and the position of the contrast agent leading edge in the next frame of the analysis image is used as the instantaneous blood flow velocity of the blood vessel segment at the position of the contrast agent leading edge in each frame of the analysis image. The sum of the differences between the instantaneous blood flow velocities of the blood vessel segment and its neighboring analyzed images in each frame of the analyzed image is calculated as an indicator of the flow velocity of the blood vessel segment at the contrast agent front position in each frame of the analyzed image. If the flow velocity judgment index is greater than the normal flow velocity threshold of the blood vessel segment, then the intersection of the contrast agent front position and the vascular skeleton line in the corresponding frame analysis image of the blood vessel segment is taken as the stenosis point of the blood vessel segment.

[0006] Further, the step of obtaining the reference fit degree between each branch point and the other branch points, and selecting a reference point for each branch point, includes: The minimum value among the lengths of the vascular skeleton line from each bifurcation point to all bifurcation points on the main vessel segment is recorded as the main vessel distance. Obtain the blood vessel diameter at each bifurcation point; randomly select two bifurcation points and record them as the example point and the analysis point, respectively. Perform negative correlation and normalization on the difference in blood vessel diameter between the example point and the analysis point and the main blood vessel distance of the analysis point, respectively, to obtain the diameter adaptation index and the distance adaptation index in turn. The preset distance weights are adjusted based on the difference in main vessel distance between the example point and the analysis point to obtain the optimized distance weights between the example point and the analysis point; the preset diameter weights and optimized distance weights are used to successively perform weighted summation on the diameter adaptation index and the distance adaptation index, and the summation result is normalized to obtain the reference adaptation degree between the example point and the analysis point; the sum of the preset diameter weights and the preset distance weights is equal to a constant 1; From the reference fit between the example point and the other bifurcation points, select the bifurcation points with reference fit greater than the preset fit threshold as the reference points of the example point.

[0007] Furthermore, the narrow points in the screening bifurcation points include: The vascular segment connected to each bifurcation point is recorded as the branch vascular segment. The position closest to the corresponding bifurcation point is selected from the contrast agent front position of the branch vascular segment in all analyzed images and recorded as the flow velocity reference position of the bifurcation point in its branch vascular segment. Calculate the difference in instantaneous blood flow velocity at the bifurcation point at the velocity reference positions of all pairs of branch vessel segments, and select the largest difference as the branch velocity difference at the bifurcation point; The difference in branch flow velocity between each bifurcation point and all its reference points is adjusted using the reference fit to obtain a narrowing judgment index for each bifurcation point; the bifurcation point corresponding to the narrowing judgment index that is greater than a preset narrowing threshold is designated as a narrowing point.

[0008] Furthermore, the grading of the degree of vascular stenosis based on the diameter of the stenosis point includes: Obtain the vessel diameter at each skeletal point on the vascular skeleton line, excluding bifurcation points; Starting from the stenosis point, a predetermined number of skeleton points are sequentially arranged along the extension direction of the blood vessel segment connected to the stenosis point to obtain the analysis sequence; the variance of the blood vessel diameter between each skeleton point and its neighboring skeleton points in the analysis sequence is calculated, and the variance is normalized to obtain the diameter fluctuation of each skeleton point. Select the first skeletal point whose diameter fluctuation is less than a preset fluctuation threshold from all the analysis sequences of the stenosis point, and calculate the mean of all stenosis point diameters as the reference diameter of the stenosis point. Based on the blood vessel diameter at the stenosis point and the reference diameter, and combined with the formula for calculating the stenosis rate, the stenosis rate at the stenosis point is obtained; based on the stenosis rate, the degree of stenosis at the location of the stenosis point is graded.

[0009] Furthermore, the method for obtaining the optimized distance weights includes: Determine whether the difference in the main blood vessel distance between the example point and the analysis point is less than a preset distance threshold. If so, use the preset distance weight as the optimized distance weight. Otherwise, perform negative correlation and normalization on the difference in the main blood vessel distance between the example point and the analysis point, and use the processing result to weight the preset distance weight to obtain the optimized distance weight.

[0010] Furthermore, the method for obtaining the narrowing judgment index includes: Calculate the absolute value of the difference between the branch flow velocity difference of each bifurcation point and all its reference points, and multiply it by the reference fit. Normalize the mean of all products to obtain the narrowing judgment index of each bifurcation point.

[0011] Furthermore, the main vascular segment is the vascular segment corresponding to the maximum value among the average vascular diameters of all vascular segments at the skeletal points.

[0012] Furthermore, the method for obtaining the normal flow rate threshold includes: The duration of contrast agent passage through the blood vessel segment is obtained, and the ratio of the length of the blood vessel segment to the duration is used as the average blood flow velocity. The standard deviation of the instantaneous blood flow velocity of all analyzed images of the blood vessel segment is calculated, and the sum of the average blood flow velocity and three times the standard deviation is used as the normal flow velocity threshold of the blood vessel segment.

[0013] Furthermore, the number of skeleton points on the vascular skeleton line within the preset neighborhood of the bifurcation point is greater than or equal to 3.

[0014] The present invention has the following beneficial effects: Firstly, due to the complexity of vascular structure, the limitations of two-dimensional projection imaging, and image ambiguity, changes in blood flow velocity can more directly reflect the degree of vascular stenosis than the narrowing ratio of the diameter or area of ​​the blood vessel in angiography images. Based on changes in blood flow velocity within the coronary arteries, the location of pathological stenosis in the middle vessels, i.e., the stenosis point, can be accurately determined.

[0015] Secondly, the special branching structure of the bifurcation point of a blood vessel causes significant changes in blood flow velocity near the bifurcation point, while the blood flow velocity within the vessel segment is relatively stable. Analyzing the coronary artery into two cases, vessel segment and bifurcation location, can improve the accuracy of locating the pathological stenosis within the coronary artery and increase the reliability of grading the degree of vascular stenosis.

[0016] Thirdly, blood flow velocity is faster at stenotic sites than in nearby areas. The pathological stenosis of a vessel segment can be located by comparing the blood flow velocity difference between the contrast agent front position in each frame of angiography and its neighboring frames. Blood flow velocity changes significantly at bifurcation points. This method determines the degree of stenosis by comparing the blood flow velocity difference between each bifurcation point and other bifurcation points with similar blood flow distribution and less hemodynamic disturbance. Because the main coronary artery is significantly affected by hemodynamic disturbances, and the vessel diameter at bifurcation points influences blood flow distribution characteristics, a comprehensive comparison is necessary. The difference in vessel diameter between the bifurcation point and other bifurcation points, as well as the distance from other bifurcation points to the main vessel segment, can measure the fit of analyzing the stenosis at the bifurcation point through other bifurcation points, and select a reference point for analyzing the stenosis at the bifurcation point. Combining the difference in blood flow velocity between the branch vessel segments of the bifurcation point and its reference point, the stenosis at the bifurcation point is determined, and the reference fit is used to measure the reliability of the stenosis. This effectively reduces the flow velocity interference and hemodynamic disturbance caused by branch structure variations and diameter differences at the bifurcation point, and reduces the risk of misjudging the stenosis at the bifurcation point. Attached Figure Description

[0017] 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.

[0018] Figure 1 This is a system structure diagram of a vascular stenosis grading system based on coronary angiography images provided in one embodiment of the present invention; Figure 2 This is a schematic diagram of an angiography image provided in one embodiment of the present invention; Figure 3 A system structure diagram for selecting a reference point for a bifurcation point, provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of a computer device for classifying the degree of vascular stenosis based on coronary angiography images, as provided in one embodiment of the present invention. Detailed Implementation

[0019] 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 vascular stenosis grading system based on coronary angiography images 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.

[0020] 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.

[0021] The following description, in conjunction with the accompanying drawings, details a specific scheme for a vascular stenosis grading system based on coronary angiography images provided by the present invention.

[0022] Example 1: Please see Figure 1 The diagram illustrates a system block diagram of a vascular stenosis grading system based on coronary angiography images according to an embodiment of the present invention. The system includes: a data acquisition module 110, an intravascular stenosis localization module 120, a bifurcation stenosis localization module 130, and a stenosis degree grading module 140.

[0023] The data acquisition module 110 is used to acquire coronary angiography video after the injection of contrast agent. The video consists of angiography images.

[0024] Coronary angiography involves injecting a contrast agent into the blood vessel being examined and capturing images of the contrast agent passing through the coronary arteries under X-ray. An angiography video consists of all frames of angiography images captured by the imaging team. The angiography video includes the dynamic process from the injection of the contrast agent into the blood vessel to the gradual filling of the entire coronary artery, clearly showing the flow trajectory and changes in filling state over time.

[0025] It should be noted that, in this embodiment of the invention, the image acquisition frequency of the multi-slice spiral CT is set to 20 frames per second to scan the coronary arteries to obtain angiographic images.

[0026] The intravascular stenosis localization module 120 is used to extract the vascular skeleton line of the vascular region in the angiography image, divide the vascular skeleton line into vascular segments based on the bifurcation point, and determine the stenosis point on each vascular segment according to the difference in blood flow velocity at the contrast agent front position in each frame of angiography image and its neighboring angiography images.

[0027] Existing methods typically classify the degree of vascular stenosis based on the narrowing ratio of the vessel diameter or area in coronary angiography images. However, due to the complexity of vascular structure, the limitations of two-dimensional projection imaging, and image ambiguity, these methods struggle to accurately pinpoint the location of pathological stenosis within the vessel. Vascular stenosis leads to accelerated blood flow, and changes in blood flow velocity more directly reflect the degree of stenosis. Therefore, this approach employs a segmented grading method based on vascular skeletonization, determining the location of stenosis in each segment based on changes in contrast agent flow velocity within the coronary artery, effectively improving the accuracy, sensitivity, and reliability of vascular stenosis grading.

[0028] The coronary arteries exhibit a tree-like branching structure. The vascular skeleton line, a single-pixel-wide centerline of the vessel, extracts the geometric and topological features of the vessels, transforming the complex vascular network into a tree-like topological structure while preserving its branching relationships and morphological characteristics. Based on the bifurcation points of the vascular skeleton line, the vessels are divided into different segments, with the two endpoints of each segment serving as bifurcation points. Dynamic monitoring of blood flow velocity within each segment can more accurately reflect the degree of stenosis in each segment, avoiding errors that may arise from overall estimation.

[0029] In normal coronary arteries, blood flow is predominantly laminar, and the contrast agent front typically advances in a smooth parabolic shape, appearing sequentially from the main coronary artery to its branches without interruption or delay. When pathological stenosis occurs, the local cross-sectional area of ​​the vessel decreases. To maintain normal blood flow, the blood velocity compensates by increasing, resulting in a faster blood flow velocity at the stenosis site compared to the areas in front of and behind it. By analyzing the difference in blood velocity at a specific location within the same vessel segment compared to its preceding and following locations, the pathological stenosis can be located, identifying the stenosis point on the vessel segment. Specifically, the change in the position of the contrast agent front on adjacent angiographic images of the vessel segment allows for analysis of blood velocity at different locations within the segment. It is important to note that this embodiment does not perform stenosis point analysis at the bifurcation points at both ends of the vessel segment.

[0030] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the vascular skeleton line includes: using the maximum inter-class variance method to obtain a segmentation threshold for the gray values ​​of pixels in the angiography image; selecting the connected regions formed by pixels with gray values ​​greater than the segmentation threshold as the vascular region; and using a double-threshold morphological thinning algorithm to thin the vascular region to obtain the vascular skeleton line. It should be noted that because the contrast agent has a significantly stronger absorption capacity for X-rays than the surrounding tissue, it appears as high brightness in the image; therefore, pixels with gray values ​​greater than the segmentation threshold are selected to constitute the vascular region. It should also be noted that since all angiography images project the same coronary artery, the vascular skeleton lines of these images are identical.

[0031] Other embodiments may also use one-dimensional maximum entropy method, histogram bimodal method, etc. for image segmentation, and use distance transformation algorithm and model-driven method for refinement operation, which will not be elaborated here.

[0032] Since a bifurcation point has at least three branches, in one implementation of this invention, the preset neighborhood is an eight-neighborhood, and the number of skeleton points on the vascular skeleton line within the preset neighborhood of the bifurcation point is greater than or equal to 3.

[0033] In one implementation of this invention, two adjacent frames of angiography images located before and after each frame of angiography image are respectively regarded as neighboring angiography images.

[0034] The bifurcation stenosis localization module 130 is used to acquire the main blood vessel segment, obtain the reference fit degree between each bifurcation point and the other bifurcation points based on the difference in blood vessel diameter between each bifurcation point and the other bifurcation points, and select the reference point for each bifurcation point; and filter the stenosis points among the bifurcation points based on the difference in blood flow velocity between each bifurcation point and its reference point and the reference fit degree.

[0035] Because blood flow is redistributed at the bifurcation of blood vessels, the difference in the number and angle of branch vessels leads to significant changes in flow velocity. At the same time, the diameter of the vessels at the branch may suddenly decrease, resulting in a compensatory increase in flow velocity. Therefore, the physiological stenosis location at the bifurcation point cannot be selected by the stenosis point location method on the vessel segment of the endovascular stenosis selection module 120.

[0036] The main coronary artery exhibits high structural stability and significant laminar flow characteristics, with uniform shear stress distribution in its wall. It is less affected by hemodynamic disturbances such as turbulence and abrupt changes in shear stress. However, the geometry of the main branch and its sub-branches at bifurcation points, such as bifurcation angle and diameter differences, can influence local hemodynamics. Therefore, the main coronary artery is significantly less affected by hemodynamic disturbances than the bifurcation or distal branch vessels. While coronary artery diameters vary at different locations, the anatomical features (diameter, blood flow distribution, etc.) at different bifurcation points with smaller diameter differences are within similar physiological ranges. Therefore, other bifurcation points with diameters close to the main coronary artery and relatively close to the main coronary artery are selected as reference points for each bifurcation point.

[0037] At bifurcation points without pathological stenosis, although the branch vessels at the bifurcation point redistribute blood flow and the angle differences between the branch vessels lead to significant changes in flow velocity, these changes are usually within the physiological adaptation range, and the difference in blood flow velocity between different branches at the bifurcation point is relatively small. In contrast, at bifurcation points with pathological stenosis, the local lumen narrowing causes a significant increase in blood flow velocity, resulting in a larger difference in blood flow velocity between different branches at the bifurcation point. Therefore, when the vessel diameter at the bifurcation point and its reference point are similar, the difference in blood flow velocity between different branches at the bifurcation point with pathological stenosis and its reference point is significant. Determining the stenosis at the bifurcation point by comparing the blood flow velocity difference between the branch vessel segments at the bifurcation point and its reference point, and using reference fit to measure the reliability of the determined stenosis, combining these two factors to screen for stenosis at the bifurcation point can effectively reduce flow velocity interference caused by branch structure variations, diameter differences, and hemodynamic disturbances at the bifurcation point, thus reducing the risk of misdiagnosis of stenosis at the bifurcation point.

[0038] In this embodiment of the invention, intravascular ultrasound technology is used to acquire cross-sectional images of coronary arteries. Continuous images are stacked to generate a three-dimensional vascular model. The bifurcation core area is obtained by radially extending 1 to 2 millimeters from the corresponding bifurcation vertex in the model as the center. Cross-sections perpendicular to the centerline are generated at 0.1 millimeter intervals along the longitudinal axis of all vessels adjacent to the bifurcation core area. The minimum value among the areas of all cross-sections in the bifurcation core area is selected as the minimum lumen area. The diameter obtained by substituting the minimum lumen area into the formula for calculating the cross-sectional area and diameter of a circular tube is used as the diameter of the vessel at the bifurcation point.

[0039] In other embodiments of the present invention, the average blood vessel diameter at the reference position of the flow velocity of each branch vessel segment is used as the blood vessel diameter of the corresponding branch vessel.

[0040] The stenosis grading module 140 is used to grade the degree of vascular stenosis based on the diameter of the stenosis point.

[0041] Based on the significant changes in blood flow velocity at the bifurcation points of coronary arteries, this study analyzes the location of physiological stenosis by dividing the vascular skeleton line into two categories: vascular segments and bifurcation points. This improves the accuracy of locating pathological stenosis in coronary arteries, thereby increasing the reliability of grading the degree of stenosis based on the vascular diameter at the stenosis point. It also accurately quantifies the severity of the lesion, providing a scientific basis for precision treatment, risk warning, and resource allocation.

[0042] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining stenosis points on a blood vessel segment includes: for each blood vessel segment, recording the angiographic images during the time period when the contrast agent flows through the blood vessel segment as analysis images; obtaining the image acquisition frequency, and using the ratio of the distance between the blood vessel segment and the position of the contrast agent leading edge in each frame of analysis image and the next frame of analysis image to the image acquisition frequency as the instantaneous blood flow velocity of the blood vessel segment at the position of the contrast agent leading edge in each frame of analysis image; calculating the sum of the differences between the instantaneous blood flow velocities of the blood vessel segment and its neighboring analysis images in each frame of analysis image as the flow velocity judgment index of the blood vessel segment at the position of the contrast agent leading edge in each frame of analysis image; if the flow velocity judgment index is greater than the normal flow velocity threshold of the blood vessel segment, then the intersection of the position of the contrast agent leading edge of the blood vessel segment and the vascular skeleton line in the corresponding frame of analysis image is taken as the stenosis point on the blood vessel segment. It should be noted that since the blood flow velocity at the narrowed location in a blood vessel is faster than the blood flow velocity in front of and behind it, the greater the difference in instantaneous blood flow velocity between the blood vessel segment in each frame of the analyzed image and its neighboring analyzed images, i.e., the greater the flow velocity judgment index, the greater the possibility that the blood vessel segment has pathological narrowing at the contrast agent leading edge position in each frame of the analyzed image. Figure 2 This is a schematic diagram of an angiography image provided in one embodiment of the present invention, as shown below. Figure 2 As shown, Figure 2 The two curves in the middle form a local blood vessel. The gray area is the part of the blood vessel filled with contrast agent. ed represents the contrast agent front. The contrast agent flows from bottom to top along the direction of the blood vessel. The contrast agent front is a line segment with a certain length.

[0043] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the normal flow velocity threshold includes: obtaining the duration of contrast agent passage through the blood vessel segment, using the ratio of the length of the blood vessel segment to the duration as the average blood flow velocity, calculating the standard deviation of the instantaneous blood flow velocity of all analyzed images of the blood vessel segment, and using the sum of the average blood flow velocity and three times the standard deviation as the normal flow velocity threshold of the blood vessel segment. It should be noted that vascular tortuosity can cause blood flow to change from laminar to turbulent or eddy flow, which may lead to fluctuations in local blood flow velocity, such as increased flow velocity at the proximal end or decreased flow velocity at the distal end. The average blood flow velocity represents the baseline of normal blood flow velocity in the blood vessel segment, the standard deviation reflects the physiological fluctuations in blood flow velocity in the blood vessel segment, and the three-times-standard-deviation principle can basically cover normal blood flow velocity fluctuations. Furthermore, the blood flow velocity is faster at the location of vascular stenosis; therefore, the normal flow velocity threshold can be used for identifying the location of vascular stenosis.

[0044] Preferably, in some possible implementations of the embodiments of the present invention, the method for selecting the reference point of the bifurcation point is described in [reference needed]. Figure 3 The diagram illustrates a system structure diagram for selecting a reference point of a bifurcation point according to an embodiment of the present invention. The structure diagram includes: an adaptation index analysis unit 131, a reference adaptation degree acquisition unit 132, and a reference point selection unit 133.

[0045] The adaptation index analysis unit 131 is used to obtain the minimum value of the vascular skeleton line length from each bifurcation point to all bifurcation points on the main vessel segment, which is recorded as the main vessel distance; obtain the vessel diameter at each bifurcation point; randomly select two bifurcation points and record them as the example point and the analysis point, respectively; perform negative correlation and normalization on the difference in vessel diameter between the example point and the analysis point and the main vessel distance of the analysis point, respectively, to obtain the diameter adaptation index and the distance adaptation index in sequence.

[0046] Calculate the average diameter of the vessel at the skeletal points on each vessel segment, and select the vessel segment with the maximum value among all vessel segments as the main vessel segment.

[0047] Although the coronary arteries as a whole have a tree-like structure, collateral vessels may be distributed in a network, resulting in multiple vascular paths from each bifurcation point to the main vessel segment, and the main vessel segment may have multiple bifurcation points. Therefore, the minimum vascular path length is selected as the distance from each bifurcation point to the main vessel. Given that other bifurcation points with similar vessel diameters and close to the main vessel are selected as reference points, the smaller the difference in vessel diameter between the example point and the analysis point, and the smaller the distance from the analysis point to the main vessel, the more suitable it is to analyze the stenosis of the example point using the difference in blood flow velocity between the branch vessel segments of the example point and the analysis point. The more suitable the vessel diameter and distance from the main vessel at the analysis point, the smaller the diameter fit index and distance fit index.

[0048] In this embodiment of the invention, the ratio of the absolute value of the difference between the vessel diameter of the example point and the analysis point to the maximum value among the absolute values ​​of the differences between the vessel diameters of all pairwise bifurcation points is calculated, and the difference between the constant 1 and the ratio is used as the diameter matching index; the ratio of the main vessel distance of the analysis point to the maximum value among the main vessel distances of all bifurcation points is calculated, and the difference between the constant 1 and the ratio is used as the distance matching index. In other embodiments of the invention, the opposite of a specific value can also be used as the exponent of an exponential function with the natural constant as the base, to achieve negative correlation and normalization of the feature values, which is not limited here.

[0049] Reference fit degree acquisition unit 132: is used to adjust the preset distance weight based on the difference in main blood vessel distance between the example point and the analysis point to obtain the optimized distance weight between the example point and the analysis point; the preset diameter weight and the optimized distance weight are used to perform weighted summation on the diameter fit index and the distance fit index in sequence, and the summation result is normalized to obtain the reference fit degree between the example point and the analysis point; the sum of the preset diameter weight and the preset distance weight is equal to the constant 1.

[0050] When the example point and the analysis point are too far apart, the reliability of analyzing the stenosis of the example point by using the difference in blood flow velocity between the branch vessel segments of the example point and the analysis point is low. It is necessary to dynamically reduce the preset distance weight based on the distance between the example point and the analysis point to avoid affecting the reliability of the stenosis analysis of the example point due to excessive distance error.

[0051] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the optimized distance weight includes: determining whether the difference in the main vessel distance between the example point and the analysis point is less than a preset distance threshold; if so, the preset distance weight is used as the optimized distance weight; otherwise, the difference in the main vessel distance between the example point and the analysis point is negatively correlated and normalized, and the preset distance weight is weighted using the processing result to obtain the optimized distance weight. It should be noted that the difference in the main vessel distance between the example point and the analysis point measures the distance between two bifurcation points. The larger the difference, the farther the distance between the two bifurcation points, and the lower the reliability of using the distance fit index to analyze the stenosis of the example point; therefore, the weight of the distance fit index should be reduced. In a specific implementation of the embodiments of the present invention, the reference fit is expressed by the formula: In the formula, The reference fit for the a-th and b-th bifurcation points; Preset diameter weights; Preset distance weights; Optimize the distance weights between the a-th and b-th bifurcation points; The diameter adaptation index for the a-th and b-th bifurcation points; The distance adaptation index for the b-th bifurcation point; This represents the absolute value of the difference in the distance between the main blood vessels at the a-th and b-th bifurcation points; The maximum absolute value of the difference between the distances from the a-th bifurcation point and the main blood vessels of the other bifurcation points is given; DY is a preset distance threshold; Norm is a normalization function. It should be noted that this embodiment uses... Achieve The negative correlation can be normalized, or function transformation can be used, which is not limited here.

[0052] In one implementation of this invention, the absolute value of the difference between the main blood vessel distances at all pairs of bifurcation points is calculated, and one-eighth of the maximum absolute value of the difference is used as a preset distance threshold.

[0053] Since the diameter of a blood vessel directly affects the stability of blood flow velocity, and changes in blood flow velocity directly reflect the degree of vascular stenosis, determining the location of stenosis based on the blood vessel diameter has a smaller error. Therefore, the blood vessel diameter is given a higher weight. However, distance matching may have errors, and there may be situations where no suitable distance exists. Therefore, the weight of the distance to the main blood vessel is reduced to prioritize diameter matching. In one implementation of this invention, the preset diameter weight is set to 0.7, and the preset distance weight is set to 0.3.

[0054] It should be noted that the reference fit of each bifurcation point with the other bifurcation points is obtained in the same way as the reference fit of the example point with the analysis point.

[0055] Reference point selection unit 133: used to select the bifurcation points with reference fit greater than the preset fit threshold from the reference fit between the example point and the other bifurcation points as the reference points of the example point.

[0056] It should be noted that a larger reference fit corresponds to a closer similarity between the bifurcation point and the example point in anatomical features such as vessel diameter and blood flow distribution. The reliability of determining the stenosis at the bifurcation point by the difference in blood flow velocity between the branch vessel segments of the example point and the bifurcation point is higher, and the bifurcation point is more likely to be a reference point for the example point.

[0057] In one implementation of this invention, the preset adaptation threshold is set to 0.8.

[0058] Preferably, in some possible implementations of the embodiments of the present invention, the method for screening stenosis points at bifurcation points includes: designating the vascular segment connected to each bifurcation point as a branch vascular segment; selecting the position closest to the corresponding bifurcation point from the contrast agent front position of the branch vascular segment in all analyzed images and designating it as the flow velocity reference position of the bifurcation point in its branch vascular segment; calculating the difference in instantaneous blood flow velocity between the bifurcation point and all its pairwise branch vascular segment flow velocity reference positions, and selecting the largest difference as the branch flow velocity difference of the bifurcation point; adjusting the difference in branch flow velocity difference between each bifurcation point and all its reference points using reference fit to obtain a stenosis judgment index for each bifurcation point; and designating the bifurcation point corresponding to a stenosis judgment index greater than a preset stenosis threshold as a stenosis point.

[0059] It should be noted that, since the proximal branch at a bifurcation point is usually thicker and the distal branch has a smaller diameter, the blood flow velocity in the proximal branch is higher than that in the distal branch. This difference in branch velocity highlights the bifurcation structure and the resulting change in blood flow velocity. A greater difference in branch velocity between the bifurcation point and the reference point, coupled with a higher reference fit, indicates a higher likelihood of pathological stenosis at the bifurcation point and a higher reliability of the stenosis assessment, thus leading to a higher stenosis assessment index. In this embodiment, the absolute value of the difference in branch velocity between each bifurcation point and all its reference points is calculated, multiplied by the reference fit. The mean of all products is then normalized to obtain the stenosis assessment index for each bifurcation point. This embodiment uses the Norm function for normalization.

[0060] In one implementation of this invention, the preset narrowing threshold is set to 0.7.

[0061] Preferably, in some possible implementations of the embodiments of the present invention, the method for classifying the degree of vascular stenosis includes: obtaining the vascular diameter of each skeleton point on the vascular skeleton line except for bifurcation points; starting from the stenosis point, sequentially arranging a predetermined number of skeleton points along the extension direction of the vascular segment connected to the stenosis point to obtain an analysis sequence; calculating the variance of the vascular diameter of each skeleton point in the analysis sequence and its neighboring skeleton points, and normalizing the variance to obtain the diameter fluctuation degree corresponding to each skeleton point; selecting the vascular diameter of the first skeleton point in all analysis sequences of the stenosis point whose diameter fluctuation degree is less than a predetermined fluctuation threshold, and calculating the mean of all vascular diameters as the reference diameter of the stenosis point; obtaining the vascular stenosis rate of the stenosis point based on the vascular stenosis rate calculation formula and the vascular diameter of the stenosis point; and classifying the degree of vascular stenosis at the location of the stenosis point based on the vascular stenosis rate. Here, the reference diameter refers to the normal vascular diameter in the vascular stenosis rate calculation formula, and the vascular stenosis rate is a well-known concept and will not be elaborated further here.

[0062] It should be noted that if the stenosis point is an internal point on the vessel segment, then skeletal points will be arranged in two opposite directions along the corresponding vessel line, starting from each stenosis point; if the stenosis point is a bifurcation point, then skeletal points will be arranged along the extension direction of the branch vessel segment at the bifurcation point, centered on the bifurcation point. The vessel diameter in the stenotic region changes gradually along the blood flow direction, specifically: starting from the stenosis initiation point, the vessel diameter gradually shrinks to the narrowest point, i.e., the most severe stenosis, and then gradually returns to the normal diameter. Although the coronary artery diameter gradually decreases along the blood flow direction, the change in vessel diameter at the stenotic location is more significant compared to the normal physiological vessel diameter change. The larger the variance of the vessel diameter between each skeletal point and its neighboring skeletal points in the analysis sequence, the greater the variation in vessel diameter between the skeletal point and its surrounding locations, and the greater the probability that the skeletal point is in a stenotic location; conversely, the greater the probability that the skeletal point is in a normal vessel location. Therefore, a diameter fluctuation less than a preset fluctuation threshold corresponds to a skeletal point location in a normal vessel. This embodiment uses max-min normalization for normalization processing; Norm functions, etc., can also be used, which will not be elaborated here.

[0063] The higher the vascular stenosis rate, the more severe the stenosis at the corresponding location in the coronary artery. In one implementation of this invention, when the vascular stenosis rate is within a certain range... At the time of stenosis, the blood vessel is only slightly narrowed at the stenosis point, and blood flow is normal; when the stenosis rate is high... At the time, moderate stenosis of the blood vessel at the stenosis point may trigger angina pectoris during physical activity; when the stenosis rate is high... At times, severe stenosis of the blood vessel at the stenosis point may lead to myocardial ischemia at rest; when the stenosis rate is within... In cases where the blood vessel at the stenosis point is severely narrowed, interventional or surgical treatment should be considered. Other embodiments can be used to classify other stenosis conditions.

[0064] In this embodiment of the invention, the preset number is set to one-tenth of the average number of skeletal points on all blood vessel segments.

[0065] In this embodiment of the invention, the two adjacent skeleton points located before and after each skeleton point in the analysis sequence are respectively regarded as neighboring skeleton points.

[0066] In this embodiment of the invention, the vascular region in the angiography image when the coronary artery is filled with contrast agent is taken as the standard vascular region. A vascular analysis line is obtained by drawing a perpendicular line from the skeletal point on the vascular segment to the vascular segment. The area within the standard vascular region that lies between the vascular analysis lines at both ends of the vascular segment and includes the vascular segment is taken as the corresponding vascular region of the vascular segment. The distance between the two intersection points of the vascular analysis line at each skeletal point on the vascular segment and the corresponding vascular region is taken as the vascular diameter of the corresponding skeletal point. It should be noted that the angiography image when the coronary artery is filled with contrast agent can clearly and accurately reflect the vascular structure.

[0067] In other embodiments of the present invention, the cross-section of the blood vessel at the corresponding point on the center line of the three-dimensional blood vessel model for each skeleton point on the blood vessel segment is obtained, and the diameter obtained by substituting the area of ​​the cross-section into the calculation formula of the cross-sectional area and diameter of the circular tube is used as the blood vessel diameter of the skeleton point on the blood vessel segment.

[0068] This invention is now complete.

[0069] Example 2: Figure 4 This is a schematic diagram of a computer device for grading the degree of vascular stenosis based on coronary angiography images, as provided in one embodiment of the present invention. For example,... Figure 4 As shown, the computer device includes: a memory 201, a processor 202, and a computer program 203 stored in the memory 201 and running on the processor 202, wherein when the processor 202 executes the computer program 203, the computer device can execute any of the aforementioned vascular stenosis grading systems based on coronary angiography images.

[0070] Furthermore, embodiments of this application also protect an apparatus that may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to execute a vascular stenosis grading system based on coronary angiography images provided in embodiments of this application.

[0071] This embodiment can divide the device into functional modules based on the above method example. For example, each module can correspond to a separate function, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0072] It should be understood that the device provided in this embodiment is used to perform the above-described vascular stenosis grading system based on coronary angiography images, and therefore can achieve the same effect as the above-described implementation method.

[0073] When using integrated units, the device may include a processing module and a storage module. When applied to a workpiece, the processing module can be used to control and manage the workpiece's operations. The storage module can be used to support the execution of program code by the workpiece.

[0074] The processing module may be a processor or a controller, which can implement or execute various exemplary logic blocks, modules, and circuits as disclosed in this application. The processor may also be a combination of computing functions, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and microprocessors, etc., and the storage module may be a memory.

[0075] Example 3: This embodiment also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer, the computer executes the above-described related method steps to implement the vascular stenosis grading system based on coronary angiography images provided in the above embodiment.

[0076] In this embodiment, the device and computer-readable storage medium are used to execute the corresponding system provided above. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects of the corresponding system provided above, and will not be repeated here.

[0077] 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.

[0078] 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.

[0079] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A vascular stenosis grading system based on coronary angiography images, characterized in that, The system includes: The data acquisition module is used to acquire coronary angiography video after the injection of contrast agent, the video consisting of angiography images; The intravascular stenosis localization module is used to extract the vascular skeleton line of the vascular region in the angiography image, divide the vascular skeleton line into vascular segments based on the bifurcation point, and determine the stenosis point on each vascular segment based on the difference in blood flow velocity at the contrast agent front position in each frame of angiography image and its neighboring angiography images. The stenosis localization module at the bifurcation point is used to acquire the main vessel segment, obtain the reference fit degree between each bifurcation point and the other bifurcation points based on the difference in vessel diameter between each bifurcation point and the other bifurcation points, and select a reference point for each bifurcation point; and filter out stenosis points among the bifurcation points based on the difference in blood flow velocity between the branch vessel segments of each bifurcation point and its reference point and the reference fit degree. The stenosis grading module is used to grade the degree of vascular stenosis based on the diameter of the stenosis point.

2. The vascular stenosis grading system based on coronary angiography images according to claim 1, characterized in that, The determination of stenosis points on each blood vessel segment includes: For each vessel segment, the angiographic images during the time period when the contrast agent flows through the vessel segment are recorded as the analysis images; The image acquisition frequency is obtained, and the ratio of the distance between the blood vessel segment and the position of the contrast agent leading edge in each frame of the analysis image and the position of the contrast agent leading edge in the next frame of the analysis image is used as the instantaneous blood flow velocity of the blood vessel segment at the position of the contrast agent leading edge in each frame of the analysis image. The sum of the differences between the instantaneous blood flow velocities of the blood vessel segment and its neighboring analyzed images in each frame of the analyzed image is calculated as an indicator of the flow velocity of the blood vessel segment at the contrast agent front position in each frame of the analyzed image. If the flow velocity judgment index is greater than the normal flow velocity threshold of the blood vessel segment, then the intersection of the contrast agent front position and the vascular skeleton line in the corresponding frame analysis image of the blood vessel segment is taken as the stenosis point of the blood vessel segment.

3. The vascular stenosis grading system based on coronary angiography images according to claim 1, characterized in that, The step of obtaining the reference fit degree between each branch point and the other branch points, and selecting the reference point for each branch point, includes: The minimum value among the lengths of the vascular skeleton line from each bifurcation point to all bifurcation points on the main vessel segment is recorded as the main vessel distance. Obtain the blood vessel diameter at each bifurcation point; randomly select two bifurcation points and record them as the example point and the analysis point, respectively. Perform negative correlation and normalization on the difference in blood vessel diameter between the example point and the analysis point and the main blood vessel distance of the analysis point, respectively, to obtain the diameter adaptation index and the distance adaptation index in turn. The preset distance weights are adjusted based on the difference in main vessel distance between the example point and the analysis point to obtain the optimized distance weights between the example point and the analysis point; the preset diameter weights and optimized distance weights are used to successively perform weighted summation on the diameter adaptation index and the distance adaptation index, and the summation result is normalized to obtain the reference adaptation degree between the example point and the analysis point; the sum of the preset diameter weights and the preset distance weights is equal to a constant 1; From the reference fit between the example point and the other bifurcation points, select the bifurcation points with reference fit greater than the preset fit threshold as the reference points of the example point.

4. A vascular stenosis grading system based on coronary angiography images according to claim 2, characterized in that, The narrow points in the screening bifurcation points include: The vascular segment connected to each bifurcation point is recorded as the branch vascular segment. The position closest to the corresponding bifurcation point is selected from the contrast agent front position of the branch vascular segment in all analyzed images and recorded as the flow velocity reference position of the bifurcation point in its branch vascular segment. Calculate the difference in instantaneous blood flow velocity at the bifurcation point at the velocity reference positions of all pairs of branch vessel segments, and select the largest difference as the branch velocity difference at the bifurcation point; The difference in branch flow velocity between each bifurcation point and all its reference points is adjusted using the reference fit to obtain a narrowing judgment index for each bifurcation point; the bifurcation point corresponding to the narrowing judgment index that is greater than a preset narrowing threshold is designated as a narrowing point.

5. A vascular stenosis grading system based on coronary angiography images according to claim 1, characterized in that, The classification of vascular stenosis degree based on the diameter of the stenosis point includes: Obtain the vessel diameter at each skeletal point on the vascular skeleton line, excluding bifurcation points; Starting from the stenosis point, a predetermined number of skeleton points are sequentially arranged along the extension direction of the blood vessel segment connected to the stenosis point to obtain the analysis sequence; the variance of the blood vessel diameter between each skeleton point and its neighboring skeleton points in the analysis sequence is calculated, and the variance is normalized to obtain the diameter fluctuation of each skeleton point. Select the first skeletal point whose diameter fluctuation is less than a preset fluctuation threshold from all the analysis sequences of the stenosis point, and calculate the mean of all stenosis point diameters as the reference diameter of the stenosis point. Based on the blood vessel diameter at the stenosis point and the reference diameter, and combined with the formula for calculating the stenosis rate, the stenosis rate at the stenosis point is obtained; based on the stenosis rate, the degree of stenosis at the location of the stenosis point is graded.

6. A vascular stenosis grading system based on coronary angiography images according to claim 3, characterized in that, The method for obtaining the optimized distance weights includes: Determine whether the difference in the main blood vessel distance between the example point and the analysis point is less than a preset distance threshold. If so, use the preset distance weight as the optimized distance weight. Otherwise, perform negative correlation and normalization on the difference in the main blood vessel distance between the example point and the analysis point, and use the processing result to weight the preset distance weight to obtain the optimized distance weight.

7. A vascular stenosis grading system based on coronary angiography images according to claim 4, characterized in that, The method for obtaining the narrowness judgment index includes: Calculate the absolute value of the difference between the branch flow velocity difference of each bifurcation point and all its reference points, and multiply it by the reference fit. Normalize the mean of all products to obtain the narrowing judgment index of each bifurcation point.

8. A vascular stenosis grading system based on coronary angiography images according to claim 1, characterized in that, The main vascular segment is the vascular segment corresponding to the maximum value among the average vascular diameters of all vascular segments and their skeletal points.

9. A vascular stenosis grading system based on coronary angiography images according to claim 2, characterized in that, The method for obtaining the normal flow rate threshold includes: The duration of contrast agent passage through the blood vessel segment is obtained, and the ratio of the length of the blood vessel segment to the duration is used as the average blood flow velocity. The standard deviation of the instantaneous blood flow velocity of all analyzed images of the blood vessel segment is calculated, and the sum of the average blood flow velocity and three times the standard deviation is used as the normal flow velocity threshold of the blood vessel segment.

10. A vascular stenosis grading system based on coronary angiography images according to claim 1, characterized in that, The number of skeleton points on the vascular skeleton line within the preset neighborhood of the bifurcation point is greater than or equal to 3.

Images