Invasive coronary angiography enhancement method and system

Abstract

The present invention relates to the field of image enhancement technology, and in particular to an invasive coronary angiography enhancement method and system, the method comprising: acquiring an invasive coronary angiography image, performing edge detection on the invasive coronary angiography image, and screening out a target edge representing the edge of a coronary artery; performing sliding window segmentation on each target edge, screening out an initial sliding window area with an edge bifurcation between the target edge; screening out suspected capillary edges and target capillary edges; screening out the capillary complete edge corresponding to each target capillary edge from the target edge image; and enhancing the target edge and the capillary complete edge in the invasive coronary angiography image to obtain a target enhanced image. The present invention realizes the enhancement of the invasive coronary angiography image, retains the detailed information of the capillaries and coronary arteries to a certain extent, thereby improving the image enhancement effect.

Description

Technical Field

[0001] The present invention relates to the technical field of image enhancement, and in particular to an invasive coronary angiography enhancement method and system. Background Art

[0002] With the development of science and technology, the application of image enhancement technology is becoming more and more extensive. For example, it can be applied to invasive coronary angiography enhancement, and can be used to enhance the characteristics of capillaries that are easily disturbed in invasive coronary angiography images to assist doctors in observation. At present, when performing image enhancement, the commonly used method is to enhance the image using grayscale histogram equalization.

[0003] However, when grayscale histogram equalization is used to enhance invasive coronary angiography images, the following technical problems often occur:

[0004] Since grayscale histogram equalization often performs statistical image enhancement according to the grayscale value distribution of the image, the grayscale level of the invasive coronary angiography image after grayscale histogram equalization often decreases, often resulting in some loss of details of capillaries, resulting in poor image enhancement effect. Summary of the invention

[0005] In order to solve the technical problem of poor image enhancement effect, the present invention proposes an invasive coronary angiography enhancement method and system.

[0006] In a first aspect, the present invention provides an invasive coronary angiography enhancement method, the method comprising:

[0007] Acquire an invasive coronary angiography image, perform edge detection on the invasive coronary angiography image to obtain a target edge image, and screen out a target edge representing a coronary artery blood vessel edge from the target edge image;

[0008] Perform sliding window segmentation on each target edge to obtain an initial sliding window region, and select the initial sliding window region with edge bifurcation between the target edge and all the initial sliding window regions as the suspected bifurcation region;

[0009] According to the angle between the bifurcation edges in the suspected bifurcation area, the suspected capillary edges are screened out from all the suspected bifurcation areas;

[0010] According to the chain code changes between the pixels on the suspected capillary edges, the target capillary edges are screened out from all the suspected capillary edges;

[0011] According to the slope change between each target capillary edge and its adjacent edge, the capillary complete edge corresponding to each target capillary edge is screened out from the target edge image;

[0012] The target edge and the complete edge of the capillary in the invasive coronary angiography image are enhanced to obtain a target enhanced image.

[0013] In combination with the first aspect above, in a possible implementation manner, screening out a target edge representing a coronary artery blood vessel edge from the target edge image includes:

[0014] Performing a closing operation on the invasive coronary angiography image to obtain a closed operation image;

[0015] Performing threshold segmentation on the closed operation image to obtain two sub-regions;

[0016] The sub-region with the lower gray value in the two sub-regions is determined as the coronary artery vessel region;

[0017] Screening out a region having the same position as the coronary artery region from the target edge image as a reference region;

[0018] The edge in each reference area is determined as the target edge.

[0019] In combination with the first aspect, in a possible implementation manner, performing sliding window segmentation on each target edge to obtain an initial sliding window area includes:

[0020] Determine any target edge as a marked edge, take any endpoint of the marked edge as a starting point, take the target length as a step length, slide a preset sliding window on the marked edge, and determine the sliding window area obtained by each sliding as an initial sliding window area, wherein the target length is the length of the preset sliding window.

[0021] In combination with the first aspect above, in a possible implementation, screening out suspected capillary edges from all suspected bifurcation regions according to the angle between bifurcation edges in the suspected bifurcation region includes:

[0022] Screen out edges branching from the target edge from all suspected bifurcation areas as reference edges, determine any reference edge as a candidate edge, and determine the target edge intersecting with the candidate edge as a temporary edge;

[0023] Determine the angle between the candidate edge and the temporary edge as the bifurcation angle corresponding to the candidate edge;

[0024] If the bifurcation angle corresponding to the candidate edge is an acute angle, the candidate edge is determined as a suspected capillary edge.

[0025] In combination with the first aspect above, in a possible implementation, screening out a target capillary edge from all suspected capillary edges according to a change in a chain code between pixels on the suspected capillary edge includes:

[0026] The suspected bifurcation area belonging to the suspected capillary edge is determined as the suspected capillary area;

[0027] Screen out suspected capillaries from the suspected capillary region based on the suspected capillary margins;

[0028] Determine the initial possibility corresponding to each suspected capillary edge according to the average width of the suspected capillaries to which each suspected capillary edge belongs, wherein the average width is negatively correlated with the initial possibility;

[0029] Obtaining the chain code value corresponding to each pixel point on each suspected capillary edge, and performing curve fitting according to the chain code values ​​corresponding to all pixels points on each suspected capillary edge to obtain the chain code curve corresponding to each suspected capillary edge;

[0030] The slope of each pixel point on the edge of each suspected capillary on its corresponding chain code curve is determined as the target slope corresponding to each pixel point on the edge of each suspected capillary;

[0031] Determine the second possibility corresponding to each suspected capillary edge according to the difference between the target slopes corresponding to all adjacent pixel points on each suspected capillary edge;

[0032] Determine the target possibility corresponding to each suspected capillary edge according to the initial possibility and the second possibility corresponding to each suspected capillary edge, wherein both the initial possibility and the second possibility are positively correlated with the target possibility;

[0033] If the target possibility corresponding to the suspected capillary edge is greater than a preset possibility threshold, the suspected capillary edge is determined as the target capillary edge.

[0034] In combination with the first aspect, in a possible implementation, determining the second possibility corresponding to each suspected capillary edge according to the difference between the target slopes corresponding to all adjacent pixels on each suspected capillary edge includes:

[0035] The accumulated value of the absolute value of the difference between the target slopes corresponding to all adjacent pixels on each suspected capillary edge is normalized to obtain the second possibility corresponding to each suspected capillary edge.

[0036] In combination with the first aspect, in a possible implementation, screening out the capillary complete edge corresponding to each target capillary edge from the target edge image according to the slope change between each target capillary edge and its adjacent edge includes:

[0037] Determine any target capillary edge as a standard edge, take the intersection of the standard edge and the target edge as an endpoint, perform curve fitting on the standard edge, and obtain a curved line corresponding to the standard edge;

[0038] Filtering out pixel points belonging to the endpoints of the standard edge from all pixel points on the curved line corresponding to the standard edge except the endpoints of the curved line as candidate pixel points;

[0039] Taking the candidate pixel point as a starting point and a preset length as a step length, sliding a target window on the curved line corresponding to the standard edge, and determining the target window area passing through the candidate pixel point as a reference area corresponding to the standard edge, wherein the preset length is a preset length of the target window;

[0040] If the target capillary edge does not exist in the reference area corresponding to the standard edge, the standard edge is determined as the complete capillary edge;

[0041] If a target capillary edge exists in the reference region corresponding to the standard edge, the standard edge and the target capillary edge in the corresponding reference region are adaptively connected to obtain a complete capillary edge corresponding to the standard edge.

[0042] In combination with the first aspect above, in a possible implementation manner, the adaptively connecting the standard edge and the target capillary edge in the corresponding reference area to obtain the complete capillary edge corresponding to the standard edge includes:

[0043] Determine each target capillary edge in the reference area corresponding to the standard edge as a possible edge to be spliced; determine the splicing possibility factor between the standard edge and each possible edge to be spliced ​​according to the slope of all pixel points on the standard edge to which it belongs and the slope of all pixel points on each possible edge to be spliced ​​on the possible edge to which it belongs;

[0044] Select the possible edge to be spliced ​​with the largest splicing possibility factor from all possible edges to be spliced ​​as the candidate splicing edge corresponding to the standard edge;

[0045] If the splicing possibility factor between the standard edge and its corresponding candidate splicing edge is less than or equal to a preset splicing threshold, the standard edge is determined as a complete capillary edge;

[0046] If the splicing possibility factor between the standard edge and its corresponding candidate splicing edge is greater than the preset splicing threshold, the standard edge and its corresponding candidate splicing edge are connected as the target splicing edge, and the standard edge is updated to the target splicing edge, and the edge splicing step is repeated until the splicing possibility factor between the latest standard edge and its corresponding candidate splicing edge is less than or equal to the preset splicing threshold, or there is no target capillary edge in the reference area corresponding to the latest standard edge, and the latest standard edge is determined as the complete capillary edge.

[0047] In combination with the first aspect above, in a possible implementation manner, a formula corresponding to a possible splicing factor between a standard edge and a possible edge to be spliced ​​is:

[0048] ;

[0049] ;in, is the possible splicing factor between the standard edge and the tth possible edge to be spliced; t is the sequence number of the possible edge to be spliced; is the normalization function; is the absolute value function; k is the mean of the slopes of all pixels on the standard edge to which they belong; It is the mean of the slopes of all pixels on the tth possible edge to be spliced; is k and The mean of It is the slope of the line between the standard edge and the tth possible edge to be spliced.

[0050] In a second aspect, the present invention provides an invasive coronary angiography enhancement system, the system comprising:

[0051] An edge detection and screening module is used to obtain an invasive coronary angiography image, perform edge detection on the invasive coronary angiography image to obtain a target edge image, and screen out a target edge representing a coronary artery vascular edge from the target edge image;

[0052] A sliding window segmentation and screening module is used to perform sliding window segmentation on each target edge to obtain an initial sliding window region, and screen out the initial sliding window region with an edge bifurcation from the target edge from all the initial sliding window regions as a suspected bifurcation region;

[0053] A suspected capillary edge screening module is used to screen suspected capillary edges from all suspected bifurcation areas according to the angle between bifurcation edges in the suspected bifurcation area;

[0054] A target capillary edge screening module is used to screen out the target capillary edge from all suspected capillary edges according to the chain code changes between the pixels on the suspected capillary edge;

[0055] A capillary complete edge screening module, used to screen out the capillary complete edge corresponding to each target capillary edge from the target edge image according to the slope change between each target capillary edge and its adjacent edge;

[0056] The edge enhancement module is used to enhance the target edge and the complete edge of the capillaries in the invasive coronary angiography image to obtain a target enhanced image.

[0057] In a third aspect, a server is provided, comprising a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, so that the device executes the method in the first aspect or any possible implementation of the first aspect.

[0058] In a fourth aspect, a computer program product is provided, comprising: a computer program code, which, when executed on a computer, enables the computer to execute the method in the first aspect or any possible implementation of the first aspect.

[0059] In a fifth aspect, a computer-readable storage medium is provided, which stores a computer program code. When the computer program code runs on a computer, the computer executes the method in the above-mentioned first aspect or any possible implementation manner of the first aspect.

[0060] The present invention has the following beneficial effects:

[0061] An invasive coronary angiography enhancement method of the present invention realizes the enhancement of invasive coronary angiography images, retains the detailed information of capillaries and coronary arteries to a certain extent, thereby improving the image enhancement effect. Compared with the use of grayscale histogram equalization to enhance the image of invasive coronary angiography images, the present invention comprehensively considers multiple factors related to capillaries, such as edge bifurcation, the angle between bifurcation edges, chain code changes, and slope changes, thereby screening out the complete edges of capillaries representing real capillaries, and enhancing the complete edges of capillaries in the invasive coronary angiography images and the target edges representing the edges of coronary arteries, retaining the detailed information of capillaries and coronary arteries to a certain extent, thereby improving the image enhancement effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0063] Figure 1 is a flow chart of an invasive coronary angiography enhancement method of the present invention;

[0064] Figure 2 It is a schematic diagram of the composition structure of an invasive coronary angiography enhancement system of the present invention;

[0065] Figure 3 The figure is a schematic diagram of the structure of a computer device of the present invention. DETAILED DESCRIPTION

[0066] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the technical solutions proposed by the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0067] Unless defined otherwise, 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 belongs.

[0068] refer to Figure 1 , showing the process of some embodiments of an invasive coronary angiography enhancement method of the present invention. The invasive coronary angiography enhancement method comprises the following steps:

[0069] Step S1, acquiring an invasive coronary angiography image, performing edge detection on the invasive coronary angiography image to obtain a target edge image, and screening out a target edge representing a coronary artery vascular edge from the target edge image.

[0070] Among them, the invasive coronary angiography image is a CT (Computed Tomography) image collected during coronary angiography. Coronary angiography is an effective method for diagnosing coronary heart disease. Coronary angiography is to insert a catheter through the femoral artery or other peripheral arteries in the thigh, send it to the ascending aorta, then find the left or right coronary artery orifice and insert it, inject iodine contrast agent to make the coronary artery visible, which can more clearly reveal the anatomical deformity of the coronary artery and the location, degree and range of its obstructive lesions.

[0071] It should be noted that the embodiment of the present invention uses a vascular contrast agent and the Judkin's method to perform angiography of the coronary arteries and obtain coronary artery images. For the captured images, artifacts caused by breathing, swallowing or internal movements such as deep inhalation are digitally enhanced by using medical image processing technology to reduce or even eliminate motion artifacts, thereby greatly improving the image quality. Among them, the Judkin's method is a commonly used coronary angiography technology, which helps doctors observe the structure and function of the coronary arteries under X-rays through specific projection technology and the use of contrast agents.

[0072] As an example, this step may include the following steps:

[0073] In the first step, invasive coronary angiography images are obtained.

[0074] For example, invasive coronary angiography images can be acquired through a CT machine.

[0075] In the second step, edge detection is performed on the invasive coronary angiography image, and the edge image obtained at this time is used as the target edge image.

[0076] The third step is to perform a closing operation on the invasive coronary angiography image to obtain a closed operation image.

[0077] The fourth step is to perform threshold segmentation on the above closed operation image to obtain two sub-regions.

[0078] In the fifth step, the sub-region with the lower gray value in the two sub-regions is determined as the coronary artery vascular region.

[0079] It should be noted that the coronary arteries often appear as thicker and darker areas in CT images.

[0080] In the sixth step, a region having the same position as the coronary artery vascular region is selected from the above target edge image as a reference region.

[0081] It should be noted that the reference region may be the region where the coronary artery vessels are located in the target edge image.

[0082] In the seventh step, the edge in each reference area is determined as the target edge.

[0083] Step S2, performing sliding window segmentation on each target edge to obtain an initial sliding window region, and screening out the initial sliding window region with an edge bifurcation between the target edge and the target edge from all the initial sliding window regions as the suspected bifurcation region.

[0084] As an example, this step may include the following steps:

[0085] In the first step, any target edge is determined as a marked edge, and a preset sliding window is slid on the marked edge with any endpoint of the marked edge as the starting point and the target length as the step length, and the sliding window area obtained by each sliding is determined as the initial sliding window area.

[0086] The target length may be the length of a preset sliding window. The preset sliding window may be a rectangular window preset for sliding. For example, the preset sliding window may be a 25×25 window. The sliding window area may be the area where the preset sliding window is located after the preset sliding window is slid.

[0087] In the second step, the initial sliding window regions with edge bifurcations between the target edge and the initial sliding window regions are selected from all the initial sliding window regions as suspected bifurcation regions.

[0088] For example, each edge except the target edge in the initial sliding window area can be determined as a non-coronary blood vessel edge. If there is a non-coronary blood vessel edge intersecting with the target edge in the initial sliding window area, the initial sliding window area can be used as a suspected bifurcation area. The initial sliding window area where there is a non-coronary blood vessel edge intersecting with the target edge is the initial sliding window area where there is an edge bifurcation with the target edge.

[0089] It should be noted that coronary arteries often generate capillaries through bifurcation. Therefore, capillaries may exist in the initial sliding window region where there is an edge bifurcation with the target edge.

[0090] Step S3, screening out suspected capillary edges from all suspected bifurcation regions according to the angles between the bifurcation edges in the suspected bifurcation regions.

[0091] The angle between the bifurcation edges is the angle between the target edge and the non-coronary blood vessel edge intersecting with it.

[0092] As an example, this step may include the following steps:

[0093] In the first step, the edges branching from the target edge are screened out from all suspected bifurcation areas as reference edges, and any reference edge is determined as a candidate edge, and the target edge intersecting with the above candidate edge is determined as a temporary edge.

[0094] Among them, the edge branched from the target edge is the non-coronary blood vessel edge intersecting with the target edge.

[0095] In the second step, the angle between the candidate edge and the temporary edge is determined as the bifurcation angle corresponding to the candidate edge.

[0096] For example, the intersection of the candidate edge and the temporary edge is determined as the target intersection, and the candidate edge is fitted with a straight line with the target intersection as the endpoint to obtain a first ray, and the temporary edge is fitted with a straight line with the target intersection as the endpoint to obtain a second ray, and the angle between the first ray and the second ray is recorded as the angle between the candidate edge and the temporary edge. The ray direction of the second ray is from the thicker part of the blood vessel represented by the temporary edge to its thinner part. The ray direction of the first ray can be from the thicker part of the area represented by the candidate edge to its thinner part.

[0097] It should be noted that the extension direction of the first ray can represent the bifurcation extension direction of the candidate edge, and the extension direction of the second ray can represent the bifurcation extension direction of the temporary edge.

[0098] Optionally, the first ray may be obtained by taking the target intersection as an endpoint and performing straight line fitting according to the positions of a preset number of pixels on the candidate edge. The ray obtained at this time may be used as the first ray. The second ray may be obtained by taking the target intersection as an endpoint and performing straight line fitting according to the positions of a preset number of pixels on the temporary edge. The ray obtained at this time may be used as the second ray. The preset number may be a preset number of pixels. For example, the preset number may be 10.

[0099] In the third step, if the bifurcation angle corresponding to the candidate edge is an acute angle, the candidate edge is determined as a suspected capillary edge.

[0100] It should be noted that coronary arteries often produce capillaries through bifurcation, and due to the influence of hemodynamic factors, the angle between the two blood vessels at the bifurcation is often less than 90°, that is, the angle is often acute, so that blood can flow more smoothly through the bifurcation instead of attaching to the blood vessel wall.

[0101] Step S4, screening out the target capillary edge from all the suspected capillary edges according to the chain code changes between the pixels on the suspected capillary edge.

[0102] As an example, this step may include the following steps:

[0103] In the first step, the suspected bifurcation area belonging to the edge of the suspected capillary is determined as the suspected capillary area.

[0104] In the second step, suspected capillaries are screened out from the suspected capillary area based on the suspected capillary edges.

[0105] The suspected capillary may be a region bounded by the edge of the suspected capillary.

[0106] In the third step, the initial possibility corresponding to each suspected capillary edge is determined according to the average width of the suspected capillaries to which each suspected capillary edge belongs.

[0107] The average width may be negatively correlated with the initial probability. The average width of the suspected capillaries may be obtained by taking the average of the widths of the suspected capillaries at different positions as the average width of the suspected capillaries.

[0108] For example, the formula for determining the initial likelihood corresponding to the edge of a suspected capillary can be:

[0109] ;in, is the initial probability corresponding to the i-th suspected capillary edge. i is the serial number of the suspected capillary edge. is a normalization function. is the average width of the suspected capillaries belonging to the edge of the i-th suspected capillary.

[0110] It should be noted that real capillaries are often thinner. The smaller it is, the more likely the suspected capillaries to which the edge of the ith suspected capillary belongs are to be real capillaries. The larger it is, the more likely it is that the suspected capillary to which the edge of the i-th suspected capillary belongs is a real capillary.

[0111] The fourth step is to obtain the chain code value corresponding to each pixel on each suspected capillary edge, and perform curve fitting according to the chain code values ​​corresponding to all pixels on each suspected capillary edge to obtain the chain code curve corresponding to each suspected capillary edge.

[0112] The chain code value corresponding to the pixel point may be the eight-chain code of the pixel point. The abscissa of the chain code curve corresponding to the suspected capillary edge may be the serial number of the pixel point on the suspected capillary edge, and the ordinate may be the chain code value corresponding to the pixel point on the suspected capillary edge.

[0113] In the fifth step, the slope of each pixel point on the edge of each suspected capillary on its corresponding chain code curve is determined as the target slope corresponding to each pixel point on the edge of each suspected capillary.

[0114] In the sixth step, the second possibility corresponding to each suspected capillary edge is determined according to the difference between the target slopes corresponding to all adjacent pixels on each suspected capillary edge.

[0115] For example, the accumulated value of the absolute values ​​of the differences between the target slopes corresponding to all adjacent pixels on each suspected capillary edge may be normalized to obtain the second possibility corresponding to each suspected capillary edge.

[0116] For example, the formula for determining the second possibility corresponding to the edge of the suspected capillary can be:

[0117] ;in, is the second possibility corresponding to the i-th suspected capillary edge. i is the serial number of the suspected capillary edge. is a normalization function. is the number of pixels on the edge of the ith suspected capillary. j is the sequence number of the pixel on the edge of the ith suspected capillary. It is the absolute value function. is the target slope corresponding to the jth pixel on the edge of the i-th suspected capillary. is the target slope corresponding to the j+1th pixel point on the edge of the i-th suspected capillary.

[0118] It should be noted that the shape of capillaries often presents certain curvature characteristics rather than being straight. Therefore, based on the curvature of the edges of suspected capillaries, the interference of straight lines on capillary identification can be reduced to a certain extent. When the value is larger, the shape of the edge of the ith suspected capillary is relatively more curved, which often means that the suspected capillary to which the edge of the ith suspected capillary belongs is more likely to be a real capillary.

[0119] In the seventh step, the target possibility corresponding to each suspected capillary edge is determined based on the initial possibility and the second possibility corresponding to each suspected capillary edge.

[0120] Among them, both the initial possibility and the second possibility can be positively correlated with the target possibility.

[0121] For example, the formula for determining the target likelihood corresponding to the edge of a suspected capillary vessel can be:

[0122] ;in, is the target probability corresponding to the i-th suspected capillary edge. i is the serial number of the suspected capillary edge. is a normalization function. is the initial probability corresponding to the i-th suspected capillary edge. is the second possibility corresponding to the i-th suspected capillary edge.

[0123] It should be noted that when The larger it is, the more likely it is that the suspected capillary to which the edge of the i-th suspected capillary belongs is a real capillary.

[0124] In the eighth step, if the target possibility corresponding to the suspected capillary edge is greater than a preset possibility threshold, the suspected capillary edge is determined as the target capillary edge.

[0125] The preset possible threshold value may be a preset threshold value for determining the edge of a capillary vessel, for example, the preset possible threshold value may be 0.75.

[0126] It should be noted that the target capillary edge is often the capillary edge located at the coronary artery, and some capillary edges may extend from the coronary artery to other locations. Therefore, the subsequent splicing of the edges to obtain the complete capillary edge can represent the real complete capillary.

[0127] Step S5, according to the slope change between each target capillary edge and its adjacent edge, the capillary complete edge corresponding to each target capillary edge is screened out from the target edge image.

[0128] As an example, this step may include the following steps:

[0129] In the first step, any target capillary edge is determined as a standard edge, and a curve fitting is performed on the standard edge with the intersection of the standard edge and the target edge as an endpoint to obtain a curve line corresponding to the standard edge.

[0130] The curved line may be a curve extending infinitely in one direction from an end point.

[0131] In the second step, the pixel points belonging to the endpoints of the above-mentioned standard edge are screened out from all the pixel points on the curved line corresponding to the above-mentioned standard edge except the endpoints of the curved line as candidate pixel points.

[0132] The third step is to slide the target window on the curved line corresponding to the standard edge with the candidate pixel point as the starting point and the preset length as the step length, and determine the target window area passing through the candidate pixel point as the reference area corresponding to the standard edge.

[0133] The preset length may be a preset length of the target window. The target window may be a square window. For example, the preset length may be 50, and the target window may be a 50×50 window. The target window area may be an area where the target window is located. The reference area corresponding to the standard edge may represent an area in the extension direction of the standard edge.

[0134] In the fourth step, if the target capillary edge does not exist in the reference area corresponding to the standard edge, the standard edge is determined as the complete capillary edge.

[0135] In the fifth step, if there is a target capillary edge in the reference area corresponding to the standard edge, the standard edge and the target capillary edge in the corresponding reference area are adaptively connected to obtain the complete capillary edge corresponding to the standard edge, which may include the following sub-steps:

[0136] In the first sub-step, each target capillary edge in the reference area corresponding to the standard edge is determined as a possible edge to be spliced.

[0137] In the second sub-step, a possible splicing factor between the standard edge and each possible edge to be spliced ​​is determined according to the slopes of all pixels on the standard edge on the standard edge to which they belong, and the slopes of all pixels on each possible edge to be spliced ​​on the possible edge to which they belong.

[0138] For example, the formula for determining the possible splicing factor between the standard edge and the possible edge to be spliced ​​can be:

[0139] ;

[0140] ;in, is the possible factor of the splicing between the standard edge and the tth possible edge to be spliced. t is the serial number of the possible edge to be spliced. is a normalization function. It is the absolute value function. k is the mean of the slopes of all pixels on the standard edge. It is the mean of the slopes of all pixels on the tth possible edge to be spliced. is k and The mean of . It is the slope of the line between the standard edge and the tth possible edge to be spliced.

[0141] It should be noted that when The smaller it is, the more similar the change trends between the standard edge and the tth possible edge to be spliced ​​are, the more likely the tth possible edge to be spliced ​​is to be an edge extended from the standard edge, and the more likely the standard edge and the tth possible edge to be spliced ​​are to belong to the same capillary. The smaller it is, the stronger the collinearity between the standard edge and the tth possible edge to be spliced. The larger it is, the more likely it is that the standard edge and the tth possible edge to be spliced ​​belong to the same capillary, and the more likely it is that the standard edge and the tth possible edge to be spliced ​​should be connected.

[0142] In the third sub-step, the possible edge to be spliced ​​with the largest splicing possibility factor is selected from all possible edges to be spliced ​​as the candidate splicing edge corresponding to the standard edge.

[0143] In the fourth sub-step, if the splicing possibility factor between the standard edge and its corresponding candidate splicing edge is less than or equal to a preset splicing threshold, the standard edge is determined as a complete capillary edge.

[0144] The preset splicing threshold may be a preset threshold, for example, the preset splicing threshold may be 0.88.

[0145] In the fifth sub-step, if the stitching possibility factor between the standard edge and its corresponding candidate stitching edge is greater than the preset stitching threshold, the standard edge and its corresponding candidate stitching edge are connected as the target stitching edge, and the standard edge is updated to the target stitching edge, and the edge stitching step is repeated until the stitching possibility factor between the latest standard edge and its corresponding candidate stitching edge is less than or equal to the preset stitching threshold, or the target capillary edge does not exist in the reference area corresponding to the latest standard edge, and the latest standard edge is determined as the complete capillary edge.

[0146] The connection between the standard edge and the corresponding candidate splicing edge can be achieved by connecting the endpoints of the standard edge and the corresponding candidate splicing edge that are close to each other. The edge splicing step may include: determining the reference area and the candidate splicing edge corresponding to the latest standard edge, and if the splicing possibility factor between the latest standard edge and the corresponding candidate splicing edge is greater than a preset splicing threshold, connecting the latest standard edge and the corresponding candidate splicing edge as the latest target splicing edge.

[0147] Step S6, enhancing the target edge and the intact edge of the capillaries in the invasive coronary angiography image to obtain a target enhanced image.

[0148] As an example, the EPST+RTV algorithm can be used to enhance the target edge and the complete edge of the capillaries in the invasive coronary angiography image, and the enhanced image obtained at this time is used as the target enhanced image. The EPST+RTV algorithm is an image enhancement algorithm based on extended phase stretch transform (EPST) and relative total variation (RTV).

[0149] refer to Figure 2Based on the same inventive concept as the above method embodiment, the present invention provides an invasive coronary angiography enhancement system, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the above computer program is executed by the processor, the steps of implementing an invasive coronary angiography enhancement method may specifically include:

[0150] The edge detection and screening module 201 is used to obtain an invasive coronary angiography image, perform edge detection on the invasive coronary angiography image to obtain a target edge image, and screen out a target edge representing a coronary artery vascular edge from the target edge image;

[0151] The sliding window segmentation and screening module 202 is used to perform sliding window segmentation on each target edge to obtain an initial sliding window region, and screen out the initial sliding window region having an edge bifurcation with the target edge from all the initial sliding window regions as a suspected bifurcation region;

[0152] A suspected capillary edge screening module 203 is used to screen suspected capillary edges from all suspected bifurcation regions according to the angle between the bifurcation edges in the suspected bifurcation regions;

[0153] The target capillary edge screening module 204 is used to screen the target capillary edge from all the suspected capillary edges according to the chain code changes between the pixels on the suspected capillary edges;

[0154] The capillary complete edge screening module 205 is used to screen out the capillary complete edge corresponding to each target capillary edge from the target edge image according to the slope change between each target capillary edge and its adjacent edge;

[0155] The edge enhancement module 206 is used to enhance the target edge and the complete edge of the capillaries in the invasive coronary angiography image to obtain a target enhanced image.

[0156] Figure 3 is a schematic diagram of the structure of a computer device provided by an embodiment of the present invention. Figure 3 As shown, the computer device 300 includes: a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302, wherein when the processor 302 executes the computer program 303, the computer device can execute any one of the invasive coronary angiography enhancement methods introduced above.

[0157] Based on the same inventive concept as the above method embodiment, the present invention provides a server, including a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, so that the device performs any of the above invasive coronary angiography enhancement methods.

[0158] Based on the same inventive concept as the above method embodiment, the present invention provides a computer program product, which includes: computer program code, when the computer program code runs on a computer, the computer executes any one of the above invasive coronary angiography enhancement methods.

[0159] Based on the same inventive concept as the above method embodiment, the present invention provides a computer-readable storage medium, which stores a computer program code. When the computer program code runs on a computer, the computer executes any one of the above invasive coronary angiography enhancement methods.

[0160] In summary, compared with the use of grayscale histogram equalization to perform image enhancement on invasive coronary angiography images, the present invention comprehensively considers multiple factors related to capillaries, such as edge bifurcation, the angle between bifurcation edges, chain code changes and slope changes, so as to screen out the complete edges of capillaries representing real capillaries, and enhance the complete edges of capillaries in invasive coronary angiography images and the target edges representing the edges of coronary arteries, thereby retaining the detailed information of capillaries and coronary arteries to a certain extent, thereby improving the image enhancement effect.

[0161] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features can be replaced by equivalents. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. An invasive coronary angiography enhancement method, characterized in that: The following steps are involved: Acquire an invasive coronary angiography image, perform edge detection on the invasive coronary angiography image to obtain a target edge image, and screen out a target edge representing a coronary artery blood vessel edge from the target edge image; Perform sliding window segmentation on each target edge to obtain an initial sliding window region, and select the initial sliding window region with edge bifurcation between the target edge and all the initial sliding window regions as the suspected bifurcation region; According to the angle between the bifurcation edges in the suspected bifurcation area, the suspected capillary edges are screened out from all the suspected bifurcation areas; According to the chain code changes between the pixels on the suspected capillary edges, the target capillary edges are screened out from all the suspected capillary edges; According to the slope change between each target capillary edge and its adjacent edge, the capillary complete edge corresponding to each target capillary edge is screened out from the target edge image; enhancing the target edge and the complete edge of the capillaries in the invasive coronary angiography image to obtain a target enhanced image; The method of screening out target capillary edges from all suspected capillary edges according to the chain code changes between pixel points on the suspected capillary edges includes: The suspected bifurcation area belonging to the suspected capillary edge is determined as the suspected capillary area; Screen out suspected capillaries from the suspected capillary region based on the suspected capillary margins; Determine the initial possibility corresponding to each suspected capillary edge according to the average width of the suspected capillaries to which each suspected capillary edge belongs, wherein the average width is negatively correlated with the initial possibility; Obtaining the chain code value corresponding to each pixel point on each suspected capillary edge, and performing curve fitting according to the chain code values ​​corresponding to all pixels points on each suspected capillary edge to obtain the chain code curve corresponding to each suspected capillary edge; The slope of each pixel point on the edge of each suspected capillary on its corresponding chain code curve is determined as the target slope corresponding to each pixel point on the edge of each suspected capillary; Determine the second possibility corresponding to each suspected capillary edge according to the difference between the target slopes corresponding to all adjacent pixel points on each suspected capillary edge; Determine the target possibility corresponding to each suspected capillary edge according to the initial possibility and the second possibility corresponding to each suspected capillary edge, wherein both the initial possibility and the second possibility are positively correlated with the target possibility; If the target possibility corresponding to the suspected capillary edge is greater than a preset possibility threshold, the suspected capillary edge is determined as the target capillary edge.

2. The invasive coronary angiography enhancement method according to claim 1, characterized in that: The step of screening out a target edge representing a coronary artery blood vessel edge from the target edge image comprises: Performing a closing operation on the invasive coronary angiography image to obtain a closed operation image; Performing threshold segmentation on the closed operation image to obtain two sub-regions; The sub-region with the lower gray value in the two sub-regions is determined as the coronary artery vessel region; Screening out a region having the same position as the coronary artery region from the target edge image as a reference region; The edge in each reference area is determined as the target edge.

3. The invasive coronary angiography enhancement method according to claim 1, characterized in that: The sliding window segmentation is performed on each target edge to obtain an initial sliding window area, including: Determine any target edge as a marked edge, take any endpoint of the marked edge as a starting point, take the target length as a step length, slide a preset sliding window on the marked edge, and determine the sliding window area obtained by each sliding as an initial sliding window area, wherein the target length is the length of the preset sliding window.

4. The invasive coronary angiography enhancement method according to claim 1, characterized in that: The method of screening out suspected capillary edges from all suspected bifurcation regions according to the angle between the bifurcation edges in the suspected bifurcation region includes: Screen out edges branching from the target edge from all suspected bifurcation areas as reference edges, determine any reference edge as a candidate edge, and determine the target edge intersecting with the candidate edge as a temporary edge; Determine the angle between the candidate edge and the temporary edge as the bifurcation angle corresponding to the candidate edge; If the bifurcation angle corresponding to the candidate edge is an acute angle, the candidate edge is determined as a suspected capillary edge.

5. The invasive coronary angiography enhancement method according to claim 1, characterized in that: The determining the second possibility corresponding to each suspected capillary edge according to the difference between the target slopes corresponding to all adjacent pixels on each suspected capillary edge includes: The accumulated value of the absolute value of the difference between the target slopes corresponding to all adjacent pixels on each suspected capillary edge is normalized to obtain the second possibility corresponding to each suspected capillary edge.

6. The invasive coronary angiography enhancement method according to claim 1, characterized in that: The step of screening out the capillary complete edge corresponding to each target capillary edge from the target edge image according to the slope change between each target capillary edge and its adjacent edge includes: Determine any target capillary edge as a standard edge, take the intersection of the standard edge and the target edge as an endpoint, perform curve fitting on the standard edge, and obtain a curved line corresponding to the standard edge; Filtering out pixel points belonging to the endpoints of the standard edge from all pixel points on the curved line corresponding to the standard edge except the endpoints of the curved line as candidate pixel points; Taking the candidate pixel point as a starting point and a preset length as a step length, sliding a target window on the curved line corresponding to the standard edge, and determining the target window area passing through the candidate pixel point as a reference area corresponding to the standard edge, wherein the preset length is a preset length of the target window; If the target capillary edge does not exist in the reference area corresponding to the standard edge, the standard edge is determined as the complete capillary edge; If a target capillary edge exists in the reference region corresponding to the standard edge, the standard edge and the target capillary edge in the corresponding reference region are adaptively connected to obtain a complete capillary edge corresponding to the standard edge.

7. The invasive coronary angiography enhancement method according to claim 6, characterized in that: The step of adaptively connecting the standard edge and the target capillary edge in the corresponding reference area to obtain the complete capillary edge corresponding to the standard edge includes: Determine each target capillary edge in the reference area corresponding to the standard edge as a possible edge to be spliced; determine the splicing possibility factor between the standard edge and each possible edge to be spliced ​​according to the slope of all pixel points on the standard edge to which it belongs and the slope of all pixel points on each possible edge to be spliced ​​on the possible edge to which it belongs; Select the possible edge to be spliced ​​with the largest splicing possibility factor from all possible edges to be spliced ​​as the candidate splicing edge corresponding to the standard edge; If the splicing possibility factor between the standard edge and its corresponding candidate splicing edge is less than or equal to a preset splicing threshold, the standard edge is determined as a complete capillary edge; If the splicing possibility factor between the standard edge and its corresponding candidate splicing edge is greater than the preset splicing threshold, the standard edge and its corresponding candidate splicing edge are connected as the target splicing edge, and the standard edge is updated to the target splicing edge, and the edge splicing step is repeated until the splicing possibility factor between the latest standard edge and its corresponding candidate splicing edge is less than or equal to the preset splicing threshold, or there is no target capillary edge in the reference area corresponding to the latest standard edge, and the latest standard edge is determined as the complete capillary edge.

8. The invasive coronary angiography enhancement method according to claim 7, characterized in that: The formula corresponding to the possible splicing factor between the standard edge and the possible edge to be spliced ​​is: ； ;in, is the possible splicing factor between the standard edge and the tth possible edge to be spliced; t is the sequence number of the possible edge to be spliced; is the normalization function; is the absolute value function; k is the mean of the slopes of all pixels on the standard edge to which they belong; It is the mean of the slopes of all pixels on the tth possible edge to be spliced; is k and The mean of It is the slope of the line between the standard edge and the tth possible edge to be spliced.

9. An invasive coronary angiography enhancement system, characterized in that: The invention comprises a processor and a memory, wherein the processor is used for processing instructions stored in the memory to implement an invasive coronary angiography enhancement method according to any one of claims 1 to 8.

Images