A method, device, equipment and medium for detecting a vascular stenosis segment

Abstract

This application provides a method, device, equipment, and medium for detecting vascular stenosis. The method includes: acquiring a binary image of a blood vessel diameter curve; determining a convex hull and multiple initial straight lines forming the convex hull in the binary image; determining multiple target straight lines from the initial straight lines, and performing line fitting on each target straight line to obtain multiple fitted straight lines and a fitted line equation corresponding to each fitted line; determining reference diameter information for each target center point in the binary image based on the fitted line equation corresponding to each fitted line; and determining at least one stenotic segment in the blood vessel and the stenosis rate corresponding to each stenotic segment using the equivalent diameter information of each target center point and the reference diameter information of each target center point. The detection method and device improve the accuracy and robustness of detecting vascular stenosis segments and stenosis rates.

Description

Technical Field

[0001] This application relates to the field of vascular detection technology, and more specifically, to a method, detection device, equipment, and medium for detecting stenotic segments of blood vessels. Background Technology

[0002] In the field of modern medical technology, vascular imaging technology can help doctors understand the condition of a patient's blood vessels. As people age, the probability of vascular problems increases, and vascular stenosis is one of the most common problems.

[0003] Currently, the detection of stenotic segments and the calculation of stenosis rates in blood vessels are generally performed by doctors who locate the stenotic segments and then calculate the stenosis rate based on visual inspection or some image measurement tools. With the increasing number of patients with vascular diseases, the doctor-patient ratio has become severely imbalanced. This method is relatively time-consuming and places a significant burden on doctors, leading to a growing clinical demand for automated solutions. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method, device, equipment and medium for detecting vascular stenosis, which is based on vascular diameter information (2D / 3D), without the need for any manual annotation or prediction model training, and makes full use of statistical, geometric and image information to perform automated and efficient calculation of vascular stenosis and stenosis rate, with higher accuracy and robustness.

[0005] In a first aspect, embodiments of this application provide a method for detecting a narrowed segment of a blood vessel, the detection method comprising:

[0006] Obtain the blood vessel diameter curve corresponding to the blood vessel; wherein, the horizontal axis of the first coordinate axis in the blood vessel diameter curve is the distance from each center point on the center line of the blood vessel to the starting point of the center line, the vertical axis of the first coordinate axis is the equivalent diameter information of each center point on the center line of the blood vessel, and the lower left corner of the blood vessel diameter curve is the origin of the first coordinate axis.

[0007] The blood vessel diameter curve is transformed to obtain a binary image of the blood vessel diameter curve corresponding to the blood vessel diameter curve; wherein, the binary image of the blood vessel diameter curve includes multiple target center points, the horizontal axis of the second coordinate axis in the binary image of the blood vessel diameter curve is the distance from each target center point to the starting point of the center line, the vertical axis of the second coordinate axis is the equivalent diameter information of each target center point, and the lower left corner of the blood vessel diameter curve is the origin of the second coordinate axis;

[0008] The convex hull is calculated using the target center point in the binary image of the blood vessel diameter curve, and the convex hull and multiple initial straight lines that make up the convex hull are determined in the binary image of the blood vessel diameter curve.

[0009] Multiple target lines are determined from the multiple initial lines, and line fitting is performed on each of the multiple target lines to obtain multiple fitted lines and the fitted line equation corresponding to each fitted line;

[0010] Based on the equation of the fitted line corresponding to each fitted line, the reference diameter information of each target center point in the binary image of the blood vessel diameter curve is determined.

[0011] By using the equivalent diameter information of each target center point and the reference diameter information of each target center point, at least one stenotic segment in the blood vessel and the stenosis rate corresponding to each stenotic segment are determined.

[0012] Furthermore, the step of performing image conversion on the blood vessel diameter curve to obtain a binary image of the blood vessel diameter curve corresponding to the blood vessel diameter curve includes:

[0013] Create a two-dimensional image with the same size as the blood vessel diameter curve and a grayscale value of 0;

[0014] Multiple target center points are determined from among the multiple center points in the blood vessel diameter curve; wherein the distance represented by the x-coordinate of the target center point is an integer;

[0015] For each target center point, determine the coordinate value of the target center point on the first coordinate axis, and use the coordinate value to add the target center point to the position corresponding to the coordinate value in the two-dimensional image;

[0016] The gray value of each target center point in the two-dimensional image is set to 1 to obtain the binary image of the blood vessel diameter curve.

[0017] Furthermore, the step of determining multiple target lines from the multiple initial lines, and performing line fitting on each of the multiple target lines to obtain multiple fitted lines and the fitted line equation corresponding to each fitted line, includes:

[0018] Based on the coordinate values ​​of multiple target center points in the binary image of the blood vessel diameter curve in the second coordinate axis, a first reference point with the smallest abscissa and a second reference point with the largest abscissa are determined from the multiple target center points, and the line connecting the first reference point and the second reference point is determined as a reference line;

[0019] Based on the reference line, multiple target lines are determined from the plurality of initial lines; wherein, the target lines are the initial lines located above or below the reference line;

[0020] For each target straight line, determine the first segment corresponding to the target straight line and multiple target center points in the first segment, and determine the start point and end point in the multiple target center points in the first segment; wherein, the start point is the target center point with the smallest x-coordinate among the multiple target center points in the first segment, and the end point is the target center point with the largest x-coordinate among the multiple target center points in the first segment;

[0021] Determine the equation of the target line corresponding to the target line, and determine the original slope parameter and the original intercept parameter in the equation of the line;

[0022] Determine the optimization function, and use the original slope parameter, the original intercept parameter, the starting point number, and the ending point number to determine the function value corresponding to the optimization function and the error value between the function value and the standard value;

[0023] If the error value is greater than or equal to the preset error threshold, the original slope parameter and the original intercept parameter are continuously adjusted until the error value between the function value corresponding to the optimization function and the standard value is less than the error threshold, thereby obtaining the target slope parameter and the target intercept parameter.

[0024] Based on the target slope parameter and the target intercept parameter, a fitted line corresponding to the target line is determined, as well as the equation of the fitted line corresponding to the fitted line.

[0025] Furthermore, after obtaining the fitted line corresponding to each target line, the detection method further includes:

[0026] For each fitted line, the average diameter of the second segment is determined by using the equivalent diameter information corresponding to each target center point in the second segment corresponding to the fitted line.

[0027] For each target center point in the second segment, calculate the diameter difference between the equivalent diameter information of the target center point and the reference diameter information of the target center point, and determine whether the diameter difference is greater than the average diameter.

[0028] If the diameter difference is greater than the average diameter, then it is determined whether the target center point is located below the fitted line;

[0029] If the target center point is located below the fitted line, then the target center point is removed.

[0030] Return to the previous steps of determining the first segment corresponding to each target line and multiple target center points in the first segment, and determining the start point and end point in the multiple target center points in the first segment, and re-perform line fitting for each target to obtain the fitted line corresponding to each target line.

[0031] Furthermore, determining at least one stenotic segment in the blood vessel using the equivalent diameter information of each target center point and the reference diameter information of each target center point includes:

[0032] The difference is calculated using the equivalent diameter information and reference diameter information of each target center point, and the difference result is binarized to obtain the difference signal; wherein, the difference signal includes multiple digital bits, and the number of digital bits in the difference signal is the same as the number of multiple target center points;

[0033] Gradient calculation is performed on the difference signal to obtain the gradient signal corresponding to the difference signal;

[0034] Based on the gradient signal, at least one stenotic segment of the blood vessel is determined.

[0035] Furthermore, determining at least one stenotic segment of the blood vessel based on the gradient signal includes:

[0036] Determine at least one starting digit bit with a value of 1 and at least one ending digit bit with a value of -1 in the gradient signal;

[0037] For each starting digital bit, the starting index of the starting digital bit in the gradient signal is determined, and the target center point corresponding to the starting index among multiple target center points is taken as the starting point of the narrow candidate segment.

[0038] In the gradient signal, determine the terminating digit to the right of the starting digit, determine the terminating digit in the gradient signal, and take the target center point corresponding to the terminating digit among the multiple target center points as the endpoint of the narrow candidate segment.

[0039] The length of the centerline of the narrow candidate segment is determined based on the distance from the starting point of the narrow candidate segment to the starting point of the centerline, and the distance from the ending point of the narrow candidate segment to the starting point of the centerline.

[0040] Determine whether the length of the centerline is less than N times the reference diameter information corresponding to any target center point in the narrow candidate segment; where N is a positive integer greater than or equal to 2;

[0041] If so, the narrow candidate segment is considered a false positive region and is excluded.

[0042] If not, the candidate stenosis segment is determined as the stenosis segment of the blood vessel.

[0043] Furthermore, after identifying at least one narrow segment, the stenosis rate corresponding to each narrow segment is determined through the following steps:

[0044] For each narrow segment, determine the reference diameter information corresponding to the starting point of the narrow segment and the reference diameter information corresponding to the ending point of the narrow segment;

[0045] Based on the reference diameter information corresponding to the starting point of the narrow segment and the reference diameter information corresponding to the ending point of the narrow segment, the reference diameter of the segment corresponding to the narrow segment is determined.

[0046] For each target center point in the narrow segment, the smallest center point with the smallest equivalent diameter information is determined among multiple target center points, and the equivalent diameter information of the smallest center point is determined as the minimum diameter of the segment corresponding to the narrow segment.

[0047] The stenosis rate corresponding to the stenosis segment is determined based on the reference diameter and minimum diameter of the segment.

[0048] Secondly, embodiments of this application also provide a device for detecting vascular stenosis, the device comprising:

[0049] The acquisition module is used to acquire the blood vessel diameter curve corresponding to the blood vessel; wherein, the horizontal axis of the first coordinate axis in the blood vessel diameter curve is the distance from each center point on the center line of the blood vessel to the starting point of the center line, the vertical axis of the first coordinate axis is the equivalent diameter information of each center point on the center line of the blood vessel, and the lower left corner of the blood vessel diameter curve is the origin of the first coordinate axis.

[0050] An image conversion module is used to convert the blood vessel diameter curve to obtain a binary image of the blood vessel diameter curve corresponding to the blood vessel diameter curve. The binary image of the blood vessel diameter curve includes multiple target center points. The horizontal axis of the second coordinate axis in the binary image of the blood vessel diameter curve is the distance from each target center point to the starting point of the center line. The vertical axis of the second coordinate axis is the equivalent diameter information of each target center point. The lower left corner of the blood vessel diameter curve is the origin of the second coordinate axis.

[0051] The convex hull calculation module is used to perform convex hull calculation using the target center point in the binary image of the blood vessel diameter curve, to determine the convex hull in the binary image of the blood vessel diameter curve, and multiple initial straight lines that make up the convex hull;

[0052] The line fitting module is used to determine multiple target lines from the multiple initial lines, and to perform line fitting on each of the multiple target lines to obtain multiple fitted lines and the fitted line equation corresponding to each fitted line.

[0053] The reference diameter information determination module is used to determine the reference diameter information of each target center point in the binary image of the blood vessel diameter curve based on the fitted line equation corresponding to each fitted line.

[0054] The stenosis segment determination module is used to determine at least one stenosis segment in the blood vessel and the stenosis rate corresponding to each stenosis segment by using the equivalent diameter information of each target center point and the reference diameter information of each target center point.

[0055] Thirdly, embodiments of this application also provide an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the method for detecting vascular stenosis as described above are performed.

[0056] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program that, when executed by a processor, performs the steps of the method for detecting vascular stenosis as described above.

[0057] The method and device for detecting vascular stenosis provided in this application, compared with the methods in the prior art, are based on vascular diameter information (2D / 3D), eliminating the need for any manual annotation and prediction model training. They fully utilize statistical, geometric, and image information, complementing each other's advantages, to perform automated, efficient, and rapid calculation of vascular stenosis segments and stenosis rates. Furthermore, they can effectively remove noise and false positives, exhibiting higher accuracy and robustness.

[0058] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0059] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0060] Figure 1A flowchart illustrating a method for detecting stenotic segments of a blood vessel, as provided in an embodiment of this application;

[0061] Figure 2 This is one of the structural schematic diagrams of a device for detecting stenotic segments of a blood vessel provided in an embodiment of this application;

[0062] Figure 3 This is a second schematic diagram of a device for detecting stenotic segments of a blood vessel, provided in an embodiment of this application.

[0063] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. Based on the embodiments of this application, every other embodiment obtained by those skilled in the art without inventive effort falls within the scope of protection of this application.

[0065] First, the applicable scenarios for this application will be introduced. This application can be applied to the field of vascular detection technology.

[0066] In the field of modern medical technology, vascular imaging technology can help doctors understand the condition of a patient's blood vessels. As people age, the probability of vascular problems increases, and vascular stenosis is one of the most common problems.

[0067] Currently, the detection of stenotic segments and the calculation of stenosis rates in blood vessels are generally performed by doctors who locate the stenotic segments and then calculate the stenosis rate based on visual inspection or some image measurement tools. With the increasing number of patients with vascular diseases, the doctor-patient ratio has become severely imbalanced. This method is relatively time-consuming and places a significant burden on doctors, leading to a growing clinical demand for automated solutions.

[0068] Research has revealed several approaches based on statistical methods for blood vessel diameter. These include least-squares fitting or curve fitting to obtain a reference diameter, detecting stenosis segments based on this reference diameter, and then calculating the stenosis rate. However, linear fitting only uses two parameters, which may not accurately reflect the actual reference diameter. Curve fitting requires pre-setting the number of parameters, and given the complex diameter variations in blood vessels, obtaining a suitable number of parameters for most situations may be difficult. Furthermore, curve fitting can easily lead to an underestimation of the stenosis rate. Other approaches employ deep learning to learn the stenosis rate. However, this requires extensive manual annotation before training the AI ​​model. Since the blood vessel diameter curve already exists, by fully utilizing other information (images, geometry, etc.), a more automated approach that eliminates the need for manual annotation and time-consuming training is more valuable for practical guidance.

[0069] People often focus on automatically obtaining blood vessel diameters, but the ability to automate the detection of stenosis and the assessment of stenosis rates after obtaining the diameter is frequently overlooked. Automating these processes and calculations can significantly accelerate vascular analysis and improve diagnostic efficiency for doctors. This is crucial for alleviating the severe doctor-patient ratio imbalance and greatly promotes integrated, fully automated diagnostic workflows. Current solutions, however, rely on a single source of information for automated calculations, which may leave some inherent limitations.

[0070] Based on this, the embodiments of this application provide a method for detecting stenotic segments of blood vessels, which improves the accuracy and robustness of detecting stenotic segments and stenosis rates of blood vessels.

[0071] Please see Figure 1 , Figure 1 This is a flowchart illustrating a method for detecting stenotic segments of a blood vessel, as provided in an embodiment of this application. Figure 1 As shown in the embodiments of this application, the method for detecting vascular stenosis includes:

[0072] S101, Obtain the blood vessel diameter curve corresponding to the blood vessel.

[0073] Here, the blood vessel diameter curve includes a first coordinate axis. The horizontal axis represents the distance from each center point on the blood vessel's centerline to the starting point of the centerline. The vertical axis represents the equivalent diameter information at each center point on the blood vessel's centerline. The lower left corner of the blood vessel diameter curve is the origin of the first coordinate axis. The equivalent diameter information refers to the actual diameter of the blood vessel.

[0074] Regarding step S101 above, in specific implementation, a blood vessel diameter curve is obtained. Specifically, the blood vessel diameter curve is generated based on the blood vessel information, which includes the centerline of the blood vessel and the equivalent diameter information of each center point on the centerline. When the blood vessel information is 2D, the equivalent diameter information is equivalent to its recorded 2D diameter. When the blood vessel information is 3D, the equivalent diameter information corresponding to its 3D blood vessel elliptical contour is calculated using the following formula:

[0075]

[0076] Among them, a i and b i Let N be the diameter of the major and minor axes of the elliptical outline of the blood vessel, and N be the number of center points of the blood vessel.

[0077] S102, perform image conversion on the blood vessel diameter curve to obtain a binary image of the blood vessel diameter curve corresponding to the blood vessel diameter curve.

[0078] Here, the binary image of the blood vessel diameter curve includes multiple target center points. These target center points are specific to a particular center point in the blood vessel diameter curve image. The binary image of the blood vessel diameter curve includes a second coordinate axis. The horizontal axis of the second coordinate axis represents the distance from each target center point to the starting point of the center line, and the vertical axis represents the equivalent diameter information of each target center point. The lower left corner of the blood vessel diameter curve image is the origin of the second coordinate axis.

[0079] In specific implementation of step S102, the blood vessel diameter curve obtained in step S101 is converted into an image to obtain a binary image of the blood vessel diameter curve corresponding to the blood vessel diameter curve.

[0080] Specifically, regarding step S102 above, the image conversion of the blood vessel diameter curve to obtain a binary image of the blood vessel diameter curve corresponding to the blood vessel diameter curve includes:

[0081] Step 1021: Create a two-dimensional image with the same size as the blood vessel diameter curve and a grayscale value of 0.

[0082] It should be noted that a two-dimensional image refers to an image with the same size as the blood vessel diameter curve and all grayscale values ​​are 0.

[0083] Regarding step 1021 above, first determine the size corresponding to the blood vessel diameter curve, then create an image of the same size as the blood vessel diameter curve, and set all gray values ​​of the image to 0 to obtain the two-dimensional image.

[0084] Step 1022: Determine multiple target center points among the multiple center points in the blood vessel diameter curve.

[0085] Here, the distance represented by the x-coordinate of the target center point is an integer.

[0086] Regarding step 1022 above, in specific implementation, for each center point in the blood vessel diameter curve, it is determined whether the distance from the center point to the starting point of the center line, represented by the horizontal coordinate of the center point, is an integer. If so, the center point is determined as the target center point. In this way, multiple target center points can be determined from multiple center points in the blood vessel diameter curve.

[0087] Step 1023: For each target center point, determine the coordinate value of the target center point on the first coordinate axis, and use the coordinate value to add the target center point to the position corresponding to the coordinate value in the two-dimensional image.

[0088] Step 1024: Set the gray value of each target center point in the two-dimensional image to 1 to obtain the binary image of the blood vessel diameter curve.

[0089] Regarding steps 1023 and 1024 above, in specific implementation, for each determined target center point, firstly, determine the coordinate value of the target center point on the first coordinate axis of the blood vessel diameter curve. Then, use the coordinate value to add the target center point to the position corresponding to the coordinate value in the created two-dimensional image. After all target center points are added to the two-dimensional image, set the grayscale value of each target center point in the two-dimensional image to 1 to obtain a black and white binary image of the blood vessel diameter curve.

[0090] S103, convex hull calculation is performed using the target center point in the binary image of the blood vessel diameter curve, and the convex hull and multiple initial straight lines that make up the convex hull are determined in the binary image of the blood vessel diameter curve.

[0091] It should be noted that a convex hull refers to a minimal convex polygon such that all target center points in the binary image of the blood vessel diameter curve lie on or within the convex hull's edges. The initial straight line is the straight line that forms this convex hull.

[0092] Regarding step S103 above, in specific implementation, convex hull calculation is performed using the target center point in the binary image of the blood vessel diameter curve. The convex hull, along with multiple initial straight lines forming the convex hull, is determined in the binary image of the blood vessel diameter curve. Specifically, during convex hull calculation, the leftmost and rightmost points in the binary image of the blood vessel diameter curve are first taken as basic points, which must lie on the convex hull. Then, the image is divided into an upper hull and a lower hull based on the line connecting these two points. For points in the upper hull, the point farthest from the line is calculated as a candidate point. Based on this candidate point and two existing points, it is calculated whether other points are on the same side of the connecting line. If so, the candidate point is determined; otherwise, the next farthest point is found as the next candidate point. The same process is applied to points in the lower hull. This continues until all candidate points are determined, ensuring that all image points are included within the convex hull. The final convex hull consists of K straight lines.

[0093] C={y=kix+bi|i=1,2,...K}.

[0094] The above method for calculating the convex hull is just an example. In practice, the method for calculating the convex hull is not limited to the example above.

[0095] S104, determine multiple target lines from the multiple initial lines, and perform line fitting on each of the multiple target lines to obtain multiple fitted lines and the fitted line equation corresponding to each fitted line.

[0096] It should be noted that the target line refers to a specific initial line determined from multiple initial lines using specific conditions. The fitted line refers to the line obtained by fitting the target line. The equation of the fitted line is the equation corresponding to the fitted line.

[0097] Regarding step S104 above, in specific implementation, multiple target lines are determined from multiple initial lines using specific conditions, and line fitting is performed on each of the multiple target lines to obtain multiple fitted lines and the fitted line equation corresponding to each fitted line.

[0098] Specifically, regarding step S104 above, determining multiple target lines from the multiple initial lines and performing line fitting on each of the multiple target lines to obtain multiple fitted lines and the fitted line equation corresponding to each fitted line includes:

[0099] Step 1041: Based on the coordinate values ​​of multiple target center points in the binary image of the blood vessel diameter curve in the second coordinate axis, determine the first reference point with the smallest abscissa and the second reference point with the largest abscissa from the multiple target center points, and determine the line connecting the first reference point and the second reference point as the reference line.

[0100] It should be noted that the first reference point refers to the target center point with the smallest x-coordinate among multiple target center points. The second reference point refers to the target center point with the largest x-coordinate among multiple target center points. The reference line is the line connecting the first and second reference points.

[0101] Regarding step 1041 above, in specific implementation, firstly, the coordinate value of each target center point in the binary image of the blood vessel diameter curve in the second coordinate axis is determined. Then, based on the coordinate value, the first reference point with the smallest abscissa and the second reference point with the largest abscissa are determined from multiple target center points. A line is then drawn between the first reference point and the second reference point, and the line is determined as the reference line.

[0102] Step 1042: Based on the reference line, determine multiple target straight lines from among the multiple initial straight lines.

[0103] It should be noted that the target line is the initial line located above or below the reference line.

[0104] Regarding step 1042 above, in practical implementation, the initial straight lines above or below the reference line cover the diameter curve, so only one type of straight line is needed here. Based on the reference line, all initial straight lines above the reference line are taken as target straight lines, or all initial straight lines below the reference line are taken as target straight lines.

[0105] Step 1043: For each target straight line, determine the first segment corresponding to the target straight line and multiple target center points in the first segment, and determine the start point and end point in the multiple target center points in the first segment.

[0106] It should be noted that the first segment refers to the range formed by the x-coordinate of the starting point of the target line and the x-coordinate of the ending point of the target line. The starting point is the target center point with the smallest x-coordinate among the multiple target center points in the first segment, and the ending point is the target center point with the largest x-coordinate among the multiple target center points in the first segment.

[0107] Regarding step 1043 above, in specific implementation, for each target straight line, the first segment corresponding to the target straight line and multiple target center points located in the first segment are determined, and the starting point and ending point are determined based on the coordinate values ​​of the multiple target center points in the first segment.

[0108] Step 1044: Determine the equation of the target line corresponding to the target line, and determine the original slope parameter and the original intercept parameter in the equation of the line.

[0109] It should be noted that the target line equation refers to the equation corresponding to the target line. The original slope parameter is the slope in the equation, and the original intercept parameter is the intercept in the equation.

[0110] Regarding step 1044 above, in specific implementation, the coordinate values ​​of any two points of the target line are used to determine the equation of the target line corresponding to the target line, and the slope in the target line equation is determined as the original slope parameter, and the intercept in the target line equation is determined as the original intercept parameter.

[0111] Step 1045: Determine the optimization function, and use the original slope parameter, the original intercept parameter, the starting point number, and the ending point number to determine the function value corresponding to the optimization function and the error value between the function value and the standard value.

[0112] It should be noted that the sequence number refers to the position number of the target center point in the binary image of the blood vessel diameter curve. For example, if there are 5 target center points in the first segment corresponding to the target straight line, which are the 6th to 10th target center points in the binary image of the blood vessel diameter curve, then the sequence number of the starting point is 6 and the sequence number of the ending point is 10. The standard value and the error value are both pre-set parameter values.

[0113] Regarding step 1045 above, in specific implementation, the optimization function to be optimized is determined, and the function value corresponding to the optimization function is determined using the original slope parameter, the original intercept parameter, the starting point number, and the ending point number. The error value is then determined using the obtained function value and a preset standard value. Specifically, the following formula is the optimization function provided in the embodiments of this application:

[0114]

[0115] Where i is the index of the target line, s i e represents the index of the starting point of the target straight line. i This indicates the index of the endpoint of the target line, Z is the function value of the optimization function, and k i b is the original slope parameter in the equation of the target line corresponding to the target line. i This is the original intercept parameter in the equation of the target line corresponding to the target line.

[0116] Step 1046: If the error value is greater than or equal to a preset error threshold, the original slope parameter and the original intercept parameter are continuously adjusted until the error value between the function value corresponding to the optimization function and the standard value is less than the error threshold, thereby obtaining the target slope parameter and the target intercept parameter.

[0117] Step 1047: Based on the target slope parameter and the target intercept parameter, determine the fitted line corresponding to the target line and the fitted line equation corresponding to the fitted line.

[0118] It should be noted that the error threshold refers to a pre-set threshold used to determine whether parameter adjustments are needed.

[0119] Regarding steps 1046 and 1047 above, in specific implementation, after calculating the error between the function value and the standard value, it is determined whether the error value is greater than or equal to a preset error threshold. If the error value is greater than or equal to the preset error threshold, the values ​​of the original slope parameter and the original intercept parameter are continuously changed, and then the function value of the optimization function is recalculated, so that the optimization function continuously decreases until the error between the function value and the standard value is less than the preset error threshold. At this point, optimization stops, and the target slope parameter and the target intercept parameter are obtained. Then, based on the target slope parameter and the target intercept parameter, the fitted line corresponding to the target line and the equation of the fitted line corresponding to the fitted line can be determined.

[0120] As an optional implementation, after obtaining the fitted line corresponding to each target line, the detection method further includes:

[0121] A: For each fitted line, the average diameter of the second segment is determined by using the equivalent diameter information of each target center point in the second segment corresponding to the fitted line.

[0122] It should be noted that the second segment refers to the range formed by the x-coordinate of the starting point of the fitted line to the x-coordinate of the ending point of the fitted line. The average diameter refers to the average diameter of all target center points in the second segment.

[0123] Regarding step A above, in specific implementation, for each fitted straight line, the second segment corresponding to the fitted straight line is determined in the binary image of the blood vessel diameter curve, as well as each target center point in the second segment. The equivalent diameter information corresponding to each target center point in the second segment is determined using the binary image of the blood vessel diameter curve, thereby determining the average diameter value corresponding to the second segment.

[0124] B: For each target center point in the second segment, calculate the diameter difference between the equivalent diameter information of the target center point and the reference diameter information of the target center point, and determine whether the diameter difference is greater than the average diameter.

[0125] C: If the diameter difference is greater than the average diameter, then determine whether the target center point is located below the fitted straight line.

[0126] D: If the target center point is located below the fitted line, then the target center point is removed.

[0127] E: Return to the steps of determining the first segment corresponding to each target line and multiple target center points in the first segment, and determining the start point and end point in the multiple target center points in the first segment, and re-perform line fitting for each target to obtain the fitted line corresponding to each target line.

[0128] Regarding steps B-E above, in practical implementation, all target center points participate in the fitting process in each second segment, which includes some noise points. Therefore, after completing the fitting process in steps 1041-1047, it is necessary to filter out those low-diameter points that may be due to stenosis to obtain the diameter of the blood vessel under healthy conditions. After determining the average diameter, for each target center point in the second segment, the diameter difference between the equivalent diameter information of the target center point and the reference diameter information of the target center point is calculated, and it is determined whether the diameter difference is greater than the average diameter. If the diameter difference is greater than the average diameter, then step C above is executed to determine whether the target center point is located below the fitted line. If the target center point is located below the fitted line, then step D above is executed to remove the target center point. Then return to step 1043 above, for each target line, determine the first segment corresponding to the target line and multiple target center points in the first segment, and determine the start point and end point in the multiple target center points in the first segment, and re-perform line fitting for each target to obtain the fitted line corresponding to each target line.

[0129] S105, Based on the fitted line equation corresponding to each fitted line, determine the reference diameter information of each target center point in the binary image of the blood vessel diameter curve.

[0130] It should be noted that the reference diameter information refers to the blood vessel reference diameter under healthy conditions.

[0131] Regarding step S105 above, in specific implementation, after obtaining multiple fitted straight lines, the reference diameter information of each target center point in the binary image of the blood vessel diameter curve can be determined based on the fitted straight line equation corresponding to each fitted straight line.

[0132] S106, using the equivalent diameter information of each target center point and the reference diameter information of each target center point, determine at least one stenotic segment in the blood vessel and the stenosis rate corresponding to each stenotic segment.

[0133] Regarding step S106 above, in specific implementation, the equivalent diameter information of each target center point can be determined using the binary image of the blood vessel diameter curve, and the reference diameter information of each target center point can be determined using the fitted straight line. Thus, using the equivalent diameter information and the reference diameter information of each target center point, at least one stenotic segment in the blood vessel and the stenosis rate corresponding to each stenotic segment can be determined.

[0134] Specifically, regarding step S106 above, determining at least one stenotic segment in the blood vessel using the equivalent diameter information of each target center point and the reference diameter information of each target center point includes:

[0135] Step 1061: Calculate the difference using the equivalent diameter information and reference diameter information of each target center point, and binarize the difference result to obtain the difference signal.

[0136] Here, the difference signal includes multiple digital bits, and the number of digital bits in the difference signal is the same as the number of target center points. For example, a binary image of a blood vessel diameter curve involves 20 target center points, meaning the number of digital bits in the difference signal is also 20.

[0137] Regarding step 1061 above, in specific implementation, after obtaining the equivalent diameter information and reference diameter information for each target center point, the difference is calculated using the equivalent diameter information and the reference diameter information, and the difference result is binarized to obtain a difference signal. Specifically, for each target center point, the difference between the equivalent diameter information and the reference diameter information of that target center point is determined, and it is determined whether the difference is less than the reference diameter information. If it is less, the digital bit corresponding to the target center point in the difference signal is set to 1; if the difference is not less than the reference diameter information, the digital bit corresponding to the target center point in the difference signal is set to 0. Target center points with a difference signal of 1 are blood vessel locations with relatively small diameters. For example, the obtained difference signal is: 00011110000111110000.

[0138] Step 1062: Perform gradient calculation on the difference signal to obtain the gradient signal corresponding to the difference signal.

[0139] Step 1063: Based on the gradient signal, determine at least one stenotic segment of the blood vessel.

[0140] Regarding steps 1062 and 1063 above, in specific implementation, after obtaining the difference signal, gradient calculation is performed on the difference signal to obtain the gradient signal corresponding to the difference signal. Specifically, in gradient calculation, the gradient signal is obtained by successively subtracting the values ​​in the difference signal. That is, the first value in the difference signal is used as the first value in the gradient signal, the difference between the second value and the first value in the difference signal is used as the second value in the gradient signal, the difference between the third value and the second value in the difference signal is used as the third value in the gradient signal, and so on, to obtain the gradient signal. Continuing the embodiment in step 1061, when the difference signal is 00011110000111110000, the obtained gradient signal is 000100-100001000-10000. After obtaining the gradient signal, at least one stenotic segment in the blood vessel can be determined.

[0141] Specifically, regarding step 1063 above, determining at least one stenotic segment of the blood vessel based on the gradient signal includes:

[0142] Step 10631: Determine at least one starting digit bit with a value of 1 and at least one ending digit bit with a value of -1 in the gradient signal.

[0143] Step 10632: For each starting digital bit, determine the starting sequence number of the starting digital bit in the gradient signal, and take the target center point corresponding to the starting sequence number among multiple target center points as the starting point of the narrow candidate segment.

[0144] Step 10633: Determine the terminating digit to the right of the starting digit in the gradient signal, determine the terminating digit in the gradient signal, and take the target center point corresponding to the terminating digit among the multiple target center points as the endpoint of the narrow candidate segment.

[0145] Regarding steps 10631-10633 above, in specific implementation, after determining the gradient signal, the position of 1 in the gradient signal is the starting point of the narrow candidate segment, and the position of -1 is the ending point of the narrow candidate segment. At least one starting digit with a value of 1 and at least one ending digit with a value of -1 are determined in the gradient signal. For each starting digit, its starting index in the gradient signal is determined, and the target center point corresponding to the starting index among multiple target center points is taken as the starting point of the narrow candidate segment. Then, the ending digit located to the right of the starting digit is determined in the gradient signal, its ending index in the gradient signal is determined, and the target center point corresponding to the ending index among multiple target center points is taken as the ending point of the narrow candidate segment.

[0146] Step 10634: Determine the length of the centerline of the narrow candidate segment based on the distance from the starting point of the narrow candidate segment to the starting point of the centerline and the distance from the ending point of the narrow candidate segment to the starting point of the centerline.

[0147] Step 10635: Determine whether the length of the centerline is less than N times the reference diameter information corresponding to any target center point in the narrow candidate segment.

[0148] Where N is a positive integer greater than or equal to 2.

[0149] Step 10636: If so, the narrow candidate segment is considered a false positive region and the narrow candidate segment is excluded.

[0150] Step 10637: If not, then the candidate stenosis segment is determined as the stenosis segment of the blood vessel.

[0151] Regarding steps 10634-10637 above, in practical implementation, since there may be some false positive areas in the initial candidate stenosis segments, it is necessary to exclude false positives. Specifically, after determining the start and end points of the candidate stenosis segment, the distance from the start point of the candidate stenosis segment to the start point of the centerline is determined based on the ordinate of the start point in the binary image of the vessel diameter curve. The distance from the end point of the candidate stenosis segment to the start point of the centerline is determined based on the ordinate of the end point in the binary image of the vessel diameter curve. Then, the centerline length of the candidate stenosis segment is determined based on the two distances. It is then determined whether the centerline length is less than N times the reference diameter information corresponding to any target center point in the candidate stenosis segment. For example, whether it is less than twice the reference diameter information. If yes, the candidate stenosis segment is considered a false positive area and is excluded. If no, the candidate stenosis segment is determined as a stenosis segment of the vessel.

[0152] As an alternative implementation method, false positives can also be eliminated in the following ways:

[0153] Calculate the ratio of the minimum diameter to the reference diameter of the narrow candidate segment. If the ratio is >0.75 (i.e., the narrowing rate is <25%), it is considered a false positive and excluded.

[0154] After screening for false positives, narrow segments with a distance less than N times the reference diameter at that location can be merged into a single narrow segment.

[0155] Specifically, after identifying at least one narrow segment, the stenosis rate corresponding to each narrow segment is determined through the following steps:

[0156] I: For each narrow segment, determine the reference diameter information corresponding to the starting point of the narrow segment and the reference diameter information corresponding to the ending point of the narrow segment.

[0157] II: Based on the reference diameter information corresponding to the starting point of the narrow segment and the reference diameter information corresponding to the ending point of the narrow segment, determine the segment reference diameter corresponding to the narrow segment.

[0158] Regarding steps I and II above, in specific implementation, for each stenotic segment, the reference diameter information corresponding to the stenosis start point and the reference diameter information corresponding to the stenosis end point are determined using the fitted straight line in the binary image of the vessel diameter curve. Specifically, given the indices of the stenosis start point and the stenosis end point, the target center points corresponding to the stenosis start point and the stenosis end point can be determined in the binary image of the vessel diameter curve, thereby determining the reference diameter information corresponding to the stenosis start point and the stenosis end point. Then, based on the reference diameter information corresponding to the stenosis start point and the stenosis end point, the segment reference diameter corresponding to the stenosis segment is determined. Specifically, the segment reference diameter is calculated using the following formula:

[0159]

[0160] III: For each target center point in the narrow segment, determine the minimum center point with the smallest equivalent diameter information among multiple target center points, and determine the equivalent diameter information of the minimum center point as the minimum diameter of the segment corresponding to the narrow segment.

[0161] Regarding step III above, in specific implementation, for each target center point in the narrow segment, firstly determine the equivalent diameter information of each target center point, then determine the smallest center point with the smallest equivalent diameter information among multiple target center points, and determine the equivalent diameter information of the smallest center point as the minimum diameter of the segment corresponding to the narrow segment.

[0162] IV: Determine the stenosis rate corresponding to the stenosis segment based on the reference diameter of the segment and the minimum diameter of the segment.

[0163] Regarding step IV above, in practical implementation, after determining the reference diameter and minimum diameter of the segment, the stenosis rate corresponding to the narrow segment can be determined based on these two diameters. Specifically, the stenosis rate of the narrow segment is calculated using the following formula:

[0164]

[0165] According to the detection method provided in the embodiments of this application, in specific implementation, after determining at least one stenotic segment and the stenosis rate corresponding to each stenotic segment, the starting point and ending point of the stenotic segment, as well as the stenosis rate of the stenotic segment, are displayed on the original vascular image for each stenotic segment. This provides doctors with an intuitive visual experience. Specifically, the distance from the starting point and ending point of the stenotic segment to the starting point of the centerline is determined based on the binary image of the vascular diameter curve. Then, the positions of the starting point and ending point can be located in the original vascular image based on the determined distance.

[0166] The method for detecting stenotic segments of blood vessels provided in this application first obtains a blood vessel diameter curve; then, it performs image conversion on the blood vessel diameter curve to obtain a binary image of the blood vessel diameter curve; it uses the target center point in the binary image of the blood vessel diameter curve to perform convex hull calculation, and determines the convex hull and multiple initial straight lines that make up the convex hull in the binary image of the blood vessel diameter curve; it determines multiple target straight lines from the multiple initial straight lines, and performs line fitting on each of the multiple target straight lines to obtain multiple fitted straight lines and the fitted straight line equation corresponding to each fitted straight line; based on the fitted straight line equation corresponding to each fitted straight line, it determines the reference diameter information of each target center point in the binary image of the blood vessel diameter curve; finally, it uses the equivalent diameter information of each target center point and the reference diameter information of each target center point to determine at least one stenotic segment in the blood vessel and the stenosis rate corresponding to each stenotic segment.

[0167] Compared with existing methods, this application is based on blood vessel diameter information (2D / 3D), which eliminates the need for any manual annotation and prediction model training. It makes full use of statistical, geometric and image information, complementing each other to perform automated and efficient calculation of vascular stenosis segments and stenosis rates. It can also effectively remove noise and false positives, and has higher accuracy and robustness.

[0168] Please see Figure 2 , Figure 2 This is one of the structural schematic diagrams of a device for detecting stenotic segments of a blood vessel provided in an embodiment of this application. Figure 2 As shown, the detection device 200 includes:

[0169] The acquisition module 201 is used to acquire the blood vessel diameter curve corresponding to the blood vessel; wherein, the horizontal axis of the first coordinate axis in the blood vessel diameter curve is the distance from each center point on the center line of the blood vessel to the starting point of the center line, the vertical axis of the first coordinate axis is the equivalent diameter information of each center point on the center line of the blood vessel, and the lower left corner of the blood vessel diameter curve is the origin of the first coordinate axis.

[0170] Image conversion module 202 is used to perform image conversion on the blood vessel diameter curve to obtain a binary image of the blood vessel diameter curve corresponding to the blood vessel diameter curve; wherein, the binary image of the blood vessel diameter curve includes multiple target center points, the horizontal axis of the second coordinate axis in the binary image of the blood vessel diameter curve is the distance from each target center point to the starting point of the center line, the vertical axis of the second coordinate axis is the equivalent diameter information of each target center point, and the lower left corner of the blood vessel diameter curve is the origin of the second coordinate axis;

[0171] The convex hull calculation module 203 is used to perform convex hull calculation using the target center point in the binary image of the blood vessel diameter curve, to determine the convex hull in the binary image of the blood vessel diameter curve, and multiple initial straight lines that make up the convex hull;

[0172] The line fitting module 204 is used to determine multiple target lines from the multiple initial lines, and to perform line fitting on each of the multiple target lines to obtain multiple fitted lines and the fitted line equation corresponding to each fitted line.

[0173] The reference diameter information determination module 205 is used to determine the reference diameter information of each target center point in the binary image of the blood vessel diameter curve based on the fitted line equation corresponding to each fitted line.

[0174] The stenosis segment determination module 206 is used to determine at least one stenosis segment in the blood vessel and the stenosis rate corresponding to each stenosis segment by using the equivalent diameter information of each target center point and the reference diameter information of each target center point.

[0175] Furthermore, when the image conversion module 202 performs image conversion on the blood vessel diameter curve to obtain a binary image of the blood vessel diameter curve corresponding to the blood vessel diameter curve, the image conversion module 202 is also used for:

[0176] Create a two-dimensional image with the same size as the blood vessel diameter curve and a grayscale value of 0;

[0177] Multiple target center points are determined from among the multiple center points in the blood vessel diameter curve; wherein the distance represented by the x-coordinate of the target center point is an integer;

[0178] For each target center point, determine the coordinate value of the target center point on the first coordinate axis, and use the coordinate value to add the target center point to the position corresponding to the coordinate value in the two-dimensional image;

[0179] The gray value of each target center point in the two-dimensional image is set to 1 to obtain the binary image of the blood vessel diameter curve.

[0180] Furthermore, when the line fitting module 204 is used to determine multiple target lines from the multiple initial lines and to perform line fitting on each of the multiple target lines to obtain multiple fitted lines and the fitted line equation corresponding to each fitted line, the line fitting module 204 is also used to:

[0181] Based on the coordinate values ​​of multiple target center points in the binary image of the blood vessel diameter curve in the second coordinate axis, a first reference point with the smallest abscissa and a second reference point with the largest abscissa are determined from the multiple target center points, and the line connecting the first reference point and the second reference point is determined as a reference line;

[0182] Based on the reference line, multiple target lines are determined from the plurality of initial lines; wherein, the target lines are the initial lines located above or below the reference line;

[0183] For each target straight line, determine the first segment corresponding to the target straight line and multiple target center points in the first segment, and determine the start point and end point in the multiple target center points in the first segment; wherein, the start point is the target center point with the smallest x-coordinate among the multiple target center points in the first segment, and the end point is the target center point with the largest x-coordinate among the multiple target center points in the first segment;

[0184] Determine the equation of the target line corresponding to the target line, and determine the original slope parameter and the original intercept parameter in the equation of the line;

[0185] Determine the optimization function, and use the original slope parameter, the original intercept parameter, the starting point number, and the ending point number to determine the function value corresponding to the optimization function and the error value between the function value and the standard value;

[0186] If the error value is greater than or equal to the preset error threshold, the original slope parameter and the original intercept parameter are continuously adjusted until the error value between the function value corresponding to the optimization function and the standard value is less than the error threshold, thereby obtaining the target slope parameter and the target intercept parameter.

[0187] Based on the target slope parameter and the target intercept parameter, a fitted line corresponding to the target line is determined, as well as the equation of the fitted line corresponding to the fitted line.

[0188] For further details, please refer to Figure 3 , Figure 3 This is a second schematic diagram of a device for detecting stenotic segments of a blood vessel, provided in an embodiment of this application. Figure 3As shown, the detection device 200 further includes a refitting module 207, which, after obtaining the fitted line corresponding to each target line, is used to:

[0189] For each fitted line, the average diameter of the second segment is determined by using the equivalent diameter information corresponding to each target center point in the second segment corresponding to the fitted line.

[0190] For each target center point in the second segment, calculate the diameter difference between the equivalent diameter information of the target center point and the reference diameter information of the target center point, and determine whether the diameter difference is greater than the average diameter.

[0191] If the diameter difference is greater than the average diameter, then it is determined whether the target center point is located below the fitted line;

[0192] If the target center point is located below the fitted line, then the target center point is removed.

[0193] Return to the previous steps of determining the first segment corresponding to each target line and multiple target center points in the first segment, and determining the start point and end point in the multiple target center points in the first segment, and re-perform line fitting for each target to obtain the fitted line corresponding to each target line.

[0194] Furthermore, when the stenosis segment determination module 206 determines at least one stenosis segment in the blood vessel using the equivalent diameter information of each target center point and the reference diameter information of each target center point, the stenosis segment determination module 206 is also used to:

[0195] The difference is calculated using the equivalent diameter information and reference diameter information of each target center point, and the difference result is binarized to obtain the difference signal; wherein, the difference signal includes multiple digital bits, and the number of digital bits in the difference signal is the same as the number of multiple target center points;

[0196] Gradient calculation is performed on the difference signal to obtain the gradient signal corresponding to the difference signal;

[0197] Based on the gradient signal, at least one stenotic segment of the blood vessel is determined.

[0198] Furthermore, when determining at least one stenotic segment of the blood vessel based on the gradient signal, the stenosis segment determination module 206 is also used to:

[0199] Determine at least one starting digit bit with a value of 1 and at least one ending digit bit with a value of -1 in the gradient signal;

[0200] For each starting digital bit, the starting index of the starting digital bit in the gradient signal is determined, and the target center point corresponding to the starting index among multiple target center points is taken as the starting point of the narrow candidate segment.

[0201] In the gradient signal, determine the terminating digit to the right of the starting digit, determine the terminating digit in the gradient signal, and take the target center point corresponding to the terminating digit among the multiple target center points as the endpoint of the narrow candidate segment.

[0202] The length of the centerline of the narrow candidate segment is determined based on the distance from the starting point of the narrow candidate segment to the starting point of the centerline, and the distance from the ending point of the narrow candidate segment to the starting point of the centerline.

[0203] Determine whether the length of the centerline is less than N times the reference diameter information corresponding to any target center point in the narrow candidate segment; where N is a positive integer greater than or equal to 2;

[0204] If so, the narrow candidate segment is considered a false positive region and is excluded.

[0205] If not, the candidate stenosis segment is determined as the stenosis segment of the blood vessel.

[0206] Furthermore, after identifying at least one narrow segment, the narrow segment identification module 206 determines the narrowing rate corresponding to each narrow segment through the following steps:

[0207] For each narrow segment, determine the reference diameter information corresponding to the starting point of the narrow segment and the reference diameter information corresponding to the ending point of the narrow segment;

[0208] Based on the reference diameter information corresponding to the starting point of the narrow segment and the reference diameter information corresponding to the ending point of the narrow segment, the reference diameter of the segment corresponding to the narrow segment is determined.

[0209] For each target center point in the narrow segment, the smallest center point with the smallest equivalent diameter information is determined among multiple target center points, and the equivalent diameter information of the smallest center point is determined as the minimum diameter of the segment corresponding to the narrow segment.

[0210] The stenosis rate corresponding to the stenosis segment is determined based on the reference diameter and minimum diameter of the segment.

[0211] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 4As shown, the electronic device 400 includes a processor 410, a memory 420, and a bus 430.

[0212] The memory 420 stores machine-readable instructions executable by the processor 410. When the electronic device 400 is running, the processor 410 communicates with the memory 420 via the bus 430. When the machine-readable instructions are executed by the processor 410, they can perform the operations described above. Figure 1 The steps of the method for detecting vascular stenosis in the illustrated method embodiment can be found in the method embodiment for specific implementation, and will not be repeated here.

[0213] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can perform the above-described actions. Figure 1 The steps of the method for detecting vascular stenosis in the illustrated method embodiment can be found in the method embodiment for specific implementation, and will not be repeated here.

[0214] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0215] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the shown or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0216] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0217] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0218] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0219] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0220] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The scope of protection of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for detecting a narrowed segment of a blood vessel, characterized in that, The detection method includes: Obtain the blood vessel diameter curve corresponding to the blood vessel; wherein, the horizontal axis of the first coordinate axis in the blood vessel diameter curve is the distance from each center point on the center line of the blood vessel to the starting point of the center line, the vertical axis of the first coordinate axis is the equivalent diameter information of each center point on the center line of the blood vessel, and the lower left corner of the blood vessel diameter curve is the origin of the first coordinate axis. The blood vessel diameter curve is transformed to obtain a binary image of the blood vessel diameter curve corresponding to the blood vessel diameter curve. The binary image of the blood vessel diameter curve includes multiple target center points. The horizontal axis of the second coordinate axis in the binary image of the blood vessel diameter curve is the distance from each target center point to the starting point of the center line. The vertical axis of the second coordinate axis is the equivalent diameter information of each target center point. The lower left corner of the blood vessel diameter curve is the origin of the second coordinate axis. The distance represented by the horizontal coordinate of the target center point is an integer. The convex hull is calculated using the target center point in the binary image of the blood vessel diameter curve, and the convex hull and multiple initial straight lines that make up the convex hull are determined in the binary image of the blood vessel diameter curve. Multiple target lines are determined from the multiple initial lines, and line fitting is performed on each of the multiple target lines to obtain multiple fitted lines and the fitted line equation corresponding to each fitted line; Based on the equation of the fitted line corresponding to each fitted line, the reference diameter information of each target center point in the binary image of the blood vessel diameter curve is determined. Using the equivalent diameter information of each target center point and the reference diameter information of each target center point, at least one stenotic segment in the blood vessel and the stenosis rate corresponding to each stenotic segment are determined; The step of determining multiple target lines from the multiple initial straight lines includes: based on the coordinate values ​​of multiple target center points in the binary image of the blood vessel diameter curve in the second coordinate axis, determining a first reference point with the smallest abscissa and a second reference point with the largest abscissa from the multiple target center points, and determining the line connecting the first reference point and the second reference point as a reference line; wherein, the first reference point refers to the target center point with the smallest abscissa among the multiple target center points, and the second reference point refers to the target center point with the largest abscissa among the multiple target center points; Based on the reference line, multiple target lines are determined from the multiple initial lines; wherein the target lines are the initial lines located above or below the reference line.

2. The detection method according to claim 1, characterized in that, The step of performing image conversion on the blood vessel diameter curve to obtain a binary image of the blood vessel diameter curve corresponding to the blood vessel diameter curve includes: Create a two-dimensional image with the same size as the blood vessel diameter curve and a grayscale value of 0; Multiple target center points are determined from among the multiple center points in the blood vessel diameter curve; wherein the distance represented by the x-coordinate of the target center point is an integer; For each target center point, determine the coordinate value of the target center point on the first coordinate axis, and use the coordinate value to add the target center point to the position corresponding to the coordinate value in the two-dimensional image; The gray value of each target center point in the two-dimensional image is set to 1 to obtain the binary image of the blood vessel diameter curve.

3. The detection method according to claim 1, characterized in that, The step of performing line fitting on each of the multiple target lines to obtain multiple fitted lines and the fitted line equation corresponding to each fitted line includes: For each target straight line, determine the first segment corresponding to the target straight line and multiple target center points in the first segment, and determine the start point and end point in the multiple target center points in the first segment; wherein, the start point is the target center point with the smallest x-coordinate among the multiple target center points in the first segment, and the end point is the target center point with the largest x-coordinate among the multiple target center points in the first segment; Determine the equation of the target line corresponding to the target line, and determine the original slope parameter and the original intercept parameter in the equation of the line; Determine the optimization function, and use the original slope parameter, the original intercept parameter, the starting point number, and the ending point number to determine the function value corresponding to the optimization function and the error value between the function value and the standard value; If the error value is greater than or equal to the preset error threshold, the original slope parameter and the original intercept parameter are continuously adjusted until the error value between the function value corresponding to the optimization function and the standard value is less than the error threshold, thereby obtaining the target slope parameter and the target intercept parameter. Based on the target slope parameter and the target intercept parameter, a fitted line corresponding to the target line is determined, as well as the equation of the fitted line corresponding to the fitted line.

4. The detection method according to claim 3, characterized in that, After obtaining the fitted line corresponding to each target line, the detection method further includes: For each fitted line, the average diameter of the second segment is determined by using the equivalent diameter information corresponding to each target center point in the second segment corresponding to the fitted line. For each target center point in the second segment, calculate the diameter difference between the equivalent diameter information of the target center point and the reference diameter information of the target center point, and determine whether the diameter difference is greater than the average diameter. If the diameter difference is greater than the average diameter, then it is determined whether the target center point is located below the fitted line; If the target center point is located below the fitted line, then the target center point is removed. Return to the previous steps of determining the first segment corresponding to each target line and multiple target center points in the first segment, and determining the start point and end point in the multiple target center points in the first segment, and re-perform line fitting for each target to obtain the fitted line corresponding to each target line.

5. The detection method according to claim 1, characterized in that, The step of determining at least one stenotic segment in the blood vessel using the equivalent diameter information of each target center point and the reference diameter information of each target center point includes: The difference is calculated using the equivalent diameter information and reference diameter information of each target center point, and the difference result is binarized to obtain the difference signal; wherein, the difference signal includes multiple digital bits, and the number of digital bits in the difference signal is the same as the number of multiple target center points; Gradient calculation is performed on the difference signal to obtain the gradient signal corresponding to the difference signal; Based on the gradient signal, at least one stenotic segment of the blood vessel is determined.

6. The detection method according to claim 5, characterized in that, Determining at least one stenotic segment of the blood vessel based on the gradient signal includes: Determine at least one starting digit bit with a value of 1 and at least one ending digit bit with a value of -1 in the gradient signal; For each starting digital bit, the starting index of the starting digital bit in the gradient signal is determined, and the target center point corresponding to the starting index among multiple target center points is taken as the starting point of the narrow candidate segment. In the gradient signal, determine the terminating digit to the right of the starting digit, determine the terminating digit in the gradient signal, and take the target center point corresponding to the terminating digit among multiple target center points as the endpoint of the narrow candidate segment. The length of the centerline of the narrow candidate segment is determined based on the distance from the starting point of the narrow candidate segment to the starting point of the centerline, and the distance from the ending point of the narrow candidate segment to the starting point of the centerline. Determine whether the length of the centerline is less than N times the reference diameter information corresponding to any target center point in the narrow candidate segment; where N is a positive integer greater than or equal to 2; If so, the narrow candidate segment is considered a false positive region and is excluded. If not, the candidate stenosis segment is determined as the stenosis segment of the blood vessel.

7. The detection method according to claim 6, characterized in that, After identifying at least one narrow segment, the stenosis rate corresponding to each narrow segment is determined through the following steps: For each narrow segment, determine the reference diameter information corresponding to the starting point of the narrow segment and the reference diameter information corresponding to the ending point of the narrow segment; Based on the reference diameter information corresponding to the starting point of the narrow segment and the reference diameter information corresponding to the ending point of the narrow segment, the reference diameter of the segment corresponding to the narrow segment is determined. For each target center point in the narrow segment, the smallest center point with the smallest equivalent diameter information is determined among multiple target center points, and the equivalent diameter information of the smallest center point is determined as the minimum diameter of the segment corresponding to the narrow segment. The stenosis rate corresponding to the stenosis segment is determined based on the reference diameter and minimum diameter of the segment.

8. A device for detecting stenotic segments of blood vessels, characterized in that, The detection device includes: The acquisition module is used to acquire the blood vessel diameter curve corresponding to the blood vessel; wherein, the horizontal axis of the first coordinate axis in the blood vessel diameter curve is the distance from each center point on the center line of the blood vessel to the starting point of the center line, the vertical axis of the first coordinate axis is the equivalent diameter information of each center point on the center line of the blood vessel, and the lower left corner of the blood vessel diameter curve is the origin of the first coordinate axis. An image conversion module is used to convert the blood vessel diameter curve to obtain a binary image of the blood vessel diameter curve corresponding to the blood vessel diameter curve. The binary image of the blood vessel diameter curve includes multiple target center points. The horizontal axis of the second coordinate axis in the binary image represents the distance from each target center point to the starting point of the center line. The vertical axis of the second coordinate axis represents the equivalent diameter information of each target center point. The lower left corner of the blood vessel diameter curve is the origin of the second coordinate axis. The distance represented by the horizontal coordinate of each target center point is an integer. The convex hull calculation module is used to perform convex hull calculation using the target center point in the binary image of the blood vessel diameter curve, to determine the convex hull in the binary image of the blood vessel diameter curve, and multiple initial straight lines that make up the convex hull; The line fitting module is used to determine multiple target lines from the multiple initial lines, and to perform line fitting on each of the multiple target lines to obtain multiple fitted lines and the fitted line equation corresponding to each fitted line. The reference diameter information determination module is used to determine the reference diameter information of each target center point in the binary image of the blood vessel diameter curve based on the fitted line equation corresponding to each fitted line. The stenosis segment determination module is used to determine at least one stenosis segment in the blood vessel and the stenosis rate corresponding to each stenosis segment by using the equivalent diameter information of each target center point and the reference diameter information of each target center point; When determining multiple target lines from the plurality of initial lines, the line fitting module is further configured to: Based on the coordinate values ​​of multiple target center points in the binary image of the blood vessel diameter curve in the second coordinate axis, a first reference point with the smallest abscissa and a second reference point with the largest abscissa are determined from the multiple target center points, and the line connecting the first reference point and the second reference point is determined as a reference line; wherein, the first reference point refers to the target center point with the smallest abscissa among the multiple target center points, and the second reference point refers to the target center point with the largest abscissa among the multiple target center points; Based on the reference line, multiple target lines are determined from the multiple initial lines; wherein the target lines are the initial lines located above or below the reference line.

9. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. The machine-readable instructions are executed by the processor to perform the steps of the method for detecting vascular stenosis as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the method for detecting vascular stenosis as described in any one of claims 1 to 7.

Images