A real-time image navigation and positioning system for cardiovascular interventional surgery

Abstract

The present invention relates to the field of cardiovascular image analysis technology, and specifically to a real-time image navigation and positioning system in cardiovascular interventional surgery. The present invention obtains the possibility of a blood vessel wall based on the grayscale difference and gradient difference between edge pixels and surrounding pixels, and then obtains blood vessel wall pixels; obtains the proximity of the blood vessel wall based on the grayscale difference and gradient difference between the blood vessel wall pixels and the surroundings, and extends to obtain extended blood vessel wall pixels; then obtains the extension direction mutation index, and screens out the direction mutation points; combines the position distribution, gradient distribution and grayscale distribution of the pixels in the moving window of the direction mutation points to obtain the abnormal possibility degree, and then obtains the abnormal circulation area; performs image enhancement on the abnormal circulation area to obtain the final cardiovascular image, and performs surgical navigation and positioning on it. The present invention improves the accuracy of real-time image navigation and positioning in cardiovascular interventional surgery.

Description

Technical Field

[0001] The present invention relates to the technical field of cardiovascular image analysis, and in particular to a real-time image navigation and positioning system in cardiovascular intervention surgery. Background Art

[0002] Cardiovascular interventional surgery is an interventional treatment method that mainly involves puncturing arteries and veins to deliver catheters to the part of the heart that needs treatment. During the operation, the position and movement trajectory of the surgical instrument in the patient's body need to be accurately navigated and positioned, so it is necessary to obtain real-time cardiovascular images of the patient's cardiovascular system for doctors to observe and plan.

[0003] In the prior art, cardiovascular images have certain resolution limitations, and the accuracy of the image needs to be enhanced by adjusting the contrast to more clearly show the difference between blood vessels and surrounding tissues. Usually, the pixels are enhanced overall or locally using algorithms such as histogram equalization and gamma correction to increase the image contrast. In reality, since the walls of coronary arteries are thin, they are easily affected by atherosclerosis, causing vascular stenosis or blockage, making the corresponding positions in cardiovascular images more difficult to identify after image enhancement, and even causing loss of details or artifacts, which greatly affects the image quality and makes it impossible to accurately navigate cardiovascular images. Summary of the invention

[0004] In order to solve the technical problem that the corresponding position in the cardiovascular image becomes more difficult to identify after image enhancement, even causes loss of details or artifacts, and thus the image quality is greatly affected, and the cardiovascular image cannot be accurately navigated, due to the thin coronary artery wall being easily affected by atherosclerosis, resulting in vascular stenosis or blockage, the object of the present invention is to provide a real-time image navigation and positioning system in cardiovascular interventional surgery, and the technical scheme adopted is as follows: a real-time image navigation and positioning system in cardiovascular interventional surgery, the system comprising: an image acquisition module, used to acquire the cardiovascular image of each patient; a mutation point recognition module, used to select an edge pixel point in the cardiovascular image as a reference pixel point, and obtain the vascular wall possibility that the reference pixel point is a vascular wall pixel point according to the grayscale difference and gradient difference between the reference pixel point and each other pixel point in a preset first neighborhood; screen all pixels according to the vascular wall possibility to obtain all vascular wall pixels in the cardiovascular image; select one vascular wall pixel point as a reference vascular wall pixel point; According to the grayscale difference and gradient difference between the reference vascular wall pixel point and each other pixel point in the preset second neighborhood, the vascular wall proximity of each other pixel point in the preset second neighborhood is obtained; according to the vascular wall proximity, the reference vascular wall pixel point is extended to obtain all extended vascular wall pixel points of the reference vascular wall pixel point; according to the change of the extension direction of each extended vascular wall pixel point, the extension direction mutation index of each extended vascular wall pixel point is obtained; according to the extension direction mutation index, the direction mutation point in the extended vascular wall pixel point is obtained; an abnormal circulation area acquisition module is used to select a pixel point as a target pixel point in the preset moving window of each direction mutation point, and obtain the abnormal possibility of the target pixel point according to the vascular wall proximity, position distribution, gradient distribution and grayscale distribution of the target pixel point; according to the abnormal possibility of all pixels, the abnormal circulation area in the cardiovascular image is obtained; a navigation and positioning module is used to perform image enhancement on the abnormal circulation area to obtain a final cardiovascular image; and surgical navigation and positioning are performed according to the final cardiovascular image.

[0005] Furthermore, the method for obtaining the possibility of the blood vessel wall includes: obtaining the possibility that the reference pixel and each other edge pixel are paired blood vessel wall pixels according to the grayscale difference and gradient direction difference between the reference pixel and other edge pixels; and obtaining the possibility of the blood vessel wall according to a blood vessel wall possibility calculation formula, and the blood vessel wall possibility calculation formula is as follows: In the formula, Represents the possibility of blood vessel wall at the reference pixel; Represents the gradient value of the reference pixel; Indicates the maximum gradient value of edge pixels in cardiovascular images; Indicates the reference pixel and the The possibility that all edge pixels are blood vessel wall pixels; Indicates the number of pixels other than the reference pixel in the preset first neighborhood; represents the normalization function; represents the maximum value function; Represents the union function.

[0006] Furthermore, a method for obtaining the paired possibility that the reference pixel and each other edge pixel are all blood vessel wall pixels includes: establishing a Cartesian coordinate system for the cardiovascular image, taking the angle between the gradient direction of the edge pixel and the positive direction of the x-axis as the gradient direction angle; and obtaining the paired possibility according to a paired possibility calculation formula, wherein the paired possibility calculation formula is as follows: In the formula, Indicates the reference pixel and the The possibility that all edge pixels are blood vessel wall pixels; Indicates the reference pixel and the The gray value of the pixel with the smaller gray value among the edge pixels; Indicates the reference pixel and the The gray value of the pixel with the larger gray value among the edge pixels; Indicates the reference pixel and the The gradient direction angle of the pixel with the larger gradient direction angle among the edge pixels; Indicates the reference pixel and the The gradient direction angle of the pixel with the smaller gradient direction angle among the edge pixels; represents the absolute value function.

[0007] Furthermore, the method for acquiring blood vessel wall pixel points includes: taking all boundary pixel points in the cardiovascular image whose blood vessel wall probability is not less than a preset first threshold as blood vessel wall pixel points.

[0008] Furthermore, the method for obtaining the blood vessel wall proximity degree includes: obtaining the blood vessel wall proximity degree according to a blood vessel wall proximity degree calculation formula, and the blood vessel wall proximity degree calculation formula is as follows: In the formula, Indicates the reference blood vessel wall pixel point preset in the second neighborhood The proximity of the other pixels to the blood vessel wall; Represents the gray value of the reference blood vessel wall pixel; Indicates the reference blood vessel wall pixel point preset in the second neighborhood Grayscale values ​​of other pixels; Represents the gradient value of the reference blood vessel wall pixel; Indicates the reference blood vessel wall pixel point preset in the second neighborhood The gradient values ​​of other pixels; Represents the gradient direction angle of the reference blood vessel wall pixel; Indicates the reference blood vessel wall pixel point preset in the second neighborhood The gradient direction angles of other pixels.

[0009] Furthermore, the method for obtaining extended blood vessel wall pixel points includes: selecting a pixel point with the greatest blood vessel wall proximity within a preset second neighborhood of a reference blood vessel wall pixel point as a first extended blood vessel wall pixel point of the reference blood vessel wall pixel point, and taking the direction from the reference blood vessel wall pixel point to the first extended blood vessel wall pixel point as the extension direction of the reference blood vessel wall pixel point, and taking the angle between the extension direction and the horizontal direction as the extension direction angle of the reference blood vessel wall pixel point; extending with the first extended blood vessel wall pixel point as the starting point until a preset number of extended blood vessel wall pixel points are obtained, and obtaining the extension direction angle of each extended blood vessel wall pixel point.

[0010] Furthermore, the method for obtaining the extension direction mutation index includes: sorting a preset number of extended blood vessel wall pixel points according to the extension order to obtain an extended blood vessel wall pixel point sequence; obtaining the extension direction mutation index according to an extension direction mutation index calculation formula, and the extension direction mutation index calculation formula is as follows: In the formula, Indicates the extension direction mutation index of the extended blood vessel wall pixel point; Represents the preset number of pixels of the extended blood vessel wall; Indicates the extension direction angle of the last extended blood vessel wall pixel point in the extended blood vessel wall pixel point sequence; Indicates the first pixel in the extended blood vessel wall sequence The extension direction angle of the extended blood vessel wall pixel.

[0011] Furthermore, the method for obtaining the direction mutation point includes: obtaining a preset second threshold, and when the ratio between the extension direction mutation index of the extended blood vessel wall pixel point and the previous extended blood vessel wall pixel point is not less than the preset second threshold, taking the extended blood vessel wall pixel point as the direction mutation point.

[0012] Furthermore, the method for obtaining the abnormal possibility degree includes: obtaining the matching degree between the extended blood vessel wall and the real blood vessel wall according to the blood vessel wall proximity degree and distance distribution between each directional mutation point and each pixel point in the preset moving window; obtaining the extended blood vessel wall pixel point according to the abnormal possibility degree calculation formula, and the abnormal possibility degree calculation formula is as follows: In the formula, Indicates the possible degree of abnormality of the target pixel; Indicates the matching degree between the extended vessel wall and the real vessel wall; Indicates the number of pixels within the preset third neighborhood of the target pixel; Indicates the third neighborhood of the target pixel. The gradient value of each pixel; Indicates the third neighborhood of the target pixel. The gray value of each pixel; Represents the standard deviation function.

[0013] Furthermore, the method for obtaining the degree of matching between the extended blood vessel wall and the real blood vessel wall includes: establishing a rectangular window with the direction mutation point as the center, moving the rectangular window in the extension direction of the previous extended blood vessel wall pixel point of the direction mutation point, and recording the trajectory of the window center during the movement as the movement trajectory; obtaining the matching degree according to the matching degree calculation formula, and the matching degree calculation formula is as follows: In the formula, Indicates the matching degree between the extended vessel wall and the real vessel wall; Indicates the number of pixels in the rectangular window during the moving process; The previous extended blood vessel wall pixel representing the direction mutation point and the first pixel in the rectangular window moving process The proximity of the blood vessel wall between pixels; Indicates the preset third neighborhood The distance between the pixel point and the moving trajectory.

[0014] The present invention has the following beneficial effects: the present invention obtains a cardiovascular image of a patient; since the stenotic and blocked areas in the cardiovascular system may lose image details or produce artifacts after cardiovascular image enhancement, the stenotic and blocked areas in the blood vessels are first identified; since the blood vessels are interconnected, the blood vessel wall pixels on the boundary of the cardiovascular image are first screened out, and the extension is performed with the pixel as the starting point; since the gradient values ​​of the blood vessel wall edge pixels are significantly different from those in other areas, and the blood vessel wall pixels often appear in pairs and have similar grayscales and opposite gradient directions, the possibility of the reference pixel being a blood vessel wall pixel is obtained based on the grayscale difference and gradient difference between the reference pixel and each other pixel in a preset first neighborhood; since the abnormal circulation phenomenon of the coronary arteries in the cardiovascular system is mostly caused by impurity accumulation, which usually does not cause actual deformation, it is necessary to analyze the proximity of the reference blood vessel wall pixel to other pixels; since the blood vessel walls on both sides of the abnormal circulation area will not Affected, it still has strong connectivity in morphology. When the reference vascular wall pixel continues to extend in the direction before the change, if it matches a vascular wall with strong connectivity within a certain distance after the extension, it means that the change may be caused by an abnormal flow area in the blood vessel; therefore, the reference vascular wall pixel is extended to obtain all extended vascular wall pixel points of the reference vascular wall pixel; according to the change of the extension direction of each extended vascular wall pixel point, the extension direction mutation index of each extended vascular wall pixel point is obtained, and then the direction mutation point is found; and according to the vascular wall proximity, position distribution, gradient distribution and grayscale distribution between the target pixel point in the preset moving window of the direction mutation point and the window center in the moving window, the possible degree of abnormality of the target pixel point is obtained; according to the possible degree of abnormality of all pixels, the abnormal flow area in the cardiovascular image is obtained; the image of the abnormal flow area is enhanced to obtain the final cardiovascular image; surgical navigation and positioning are performed based on the final cardiovascular image.

[0015] The present invention can enhance the abnormal circulation area and the normal area of ​​the cardiovascular image to different degrees, so that the enhanced final cardiovascular image can better display the vascular structure, thereby improving the accuracy of real-time image navigation and positioning in cardiovascular interventional surgery. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] 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.

[0017] Figure 1A module diagram of a real-time image navigation and positioning system for cardiovascular interventional surgery provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation, structure, features and effects of a real-time image navigation and positioning system in cardiovascular intervention surgery proposed by the present invention in combination 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.

[0019] 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.

[0020] The following is a detailed description of a specific scheme of a real-time image navigation and positioning system for cardiovascular interventional surgery provided by the present invention in conjunction with the accompanying drawings.

[0021] See also Figure 1 , which shows a real-time image navigation and positioning system in cardiovascular intervention surgery provided by an embodiment of the present invention, the system includes: an image acquisition module 101, a mutation point recognition module 102, an abnormal circulation area acquisition module 103 and a navigation and positioning module 104.

[0022] The image acquisition module 101 is used to acquire cardiovascular images of each patient.

[0023] The embodiment of the present invention is mainly applied to the scenario of locating cardiovascular blockage. In order to locate the cardiovascular blockage area, a cardiovascular image of the patient is first acquired, and then subsequent image analysis is performed.

[0024] In one embodiment of the present invention, a DSA device is used to obtain a digital subtraction angiography image as a cardiovascular image for subsequent analysis. It should be noted that in addition to the digital subtraction angiography image, a CT image can also be used for subsequent analysis. The method of obtaining a digital subtraction angiography image or a CT image is a technical means well known to those skilled in the art and is not limited or elaborated herein.

[0025] The mutation point identification module 102 is used to select any edge pixel point in the cardiovascular image as a reference pixel point, and obtain the vascular wall possibility that the reference pixel point is a vascular wall pixel point according to the grayscale difference and gradient difference between the reference pixel point and each other pixel point in a preset first neighborhood; screen all pixel points according to the vascular wall possibility to obtain all vascular wall pixel points in the cardiovascular image; select any vascular wall pixel point as a reference vascular wall pixel point; obtain the vascular wall proximity of each other pixel point in the preset second neighborhood according to the grayscale difference and gradient difference between the reference vascular wall pixel point and each other pixel point in a preset second neighborhood; extend the reference vascular wall pixel point according to the vascular wall proximity to obtain all extended vascular wall pixel points of the reference vascular wall pixel point; obtain an extension direction mutation index of each extended vascular wall pixel point according to the extension direction change of each extended vascular wall pixel point; and obtain a direction mutation point in the extended vascular wall pixel point according to the extension direction mutation index.

[0026] In reality, the coronary arteries in the cardiovascular system are prone to vascular stenosis or blockage due to thrombosis, plaque deposition, inflammation, etc., resulting in poor blood flow or even occlusion. The abnormal blood flow in these locations will lead to uneven distribution of contrast agents, resulting in poor imaging quality of digital subtraction angiography images. In addition, the abnormal area will be more difficult to identify after the overall image enhancement, and image details may be lost or artifacts may be generated. Therefore, in order to avoid the degradation of image quality after enhancement, it is necessary to first identify the stenosis and blockage areas in the blood vessels. Since all blood vessels in cardiovascular images are interconnected, and the blood vessels extend from the image boundary to the inside. Therefore, we first screen out the vascular wall pixels on the boundary of the cardiovascular image, and extend inward from the pixel point as the starting point, so as to identify the stenosis and blockage areas according to the changes in the vascular wall during the extension process. In reality, there is a large difference between the gradient values ​​of the edge pixels of the blood vessel wall and those of other areas, and the blood vessel wall pixels often appear in pairs and have similar grayscales and opposite gradient directions. Therefore, in an embodiment of the present invention, the possibility that the reference pixel is a blood vessel wall pixel is obtained based on the grayscale difference and gradient difference between the reference pixel and each other pixel in the preset first neighborhood.

[0027] Preferably, in one embodiment of the present invention, a method for obtaining the possibility of a blood vessel wall includes: in reality, blood vessel wall pixels usually appear in pairs and have similar grayscale values ​​and opposite gradient directions. Therefore, in an embodiment of the present invention, based on the grayscale difference and gradient direction difference between the reference pixel and other edge pixels, the possibility that the reference pixel and each other edge pixel are all blood vessel wall pixels is obtained. The specific method includes: establishing a Cartesian coordinate system for the cardiovascular image, taking the angle between the gradient direction of the edge pixel and the positive direction of the x-axis as the gradient direction angle; obtaining the paired possibility according to a paired possibility calculation formula, and the paired possibility calculation formula is as follows: In the formula, Indicates the reference pixel and the The possibility that all edge pixels are blood vessel wall pixels; Indicates the reference pixel and the The gray value of the pixel with the smaller gray value among the edge pixels; Indicates the reference pixel and the The gray value of the pixel with the larger gray value among the edge pixels; Indicates the reference pixel and the The gradient direction angle of the pixel with the larger gradient direction angle among the edge pixels; Indicates the reference pixel and the The gradient direction angle of the pixel with the smaller gradient direction angle among the edge pixels; represents the absolute value function.

[0028] In the pairwise probability calculation formula, when the larger gray value of the edge pixel With smaller gray value The smaller the difference between The smaller the value, the greater the possibility that the pixel points corresponding to the above two gray values ​​are paired with the edge points of the blood vessel wall; when the larger gradient direction of the edge pixel point is closer to the opposite direction of the smaller gradient direction, The closer it is to 0, the greater the possibility of pairing between corresponding pixels.

[0029] The vascular wall probability is obtained according to the vascular wall probability calculation formula, which is as follows: In the formula, Represents the possibility of blood vessel wall at the reference pixel; Represents the gradient value of the reference pixel; Indicates the maximum gradient value of edge pixels in cardiovascular images; Indicates the reference pixel and the The possibility that all edge pixels are blood vessel wall pixels; Indicates the number of pixels other than the reference pixel in the preset first neighborhood; represents the normalization function; represents the maximum value function; Represents the union function.

[0030] In the vascular wall probability calculation formula, the closer the reference pixel is to the maximum gradient value, the more likely it is that the reference pixel is a vascular wall pixel; if the maximum value of the pairing probability between the reference pixel and other edge pixels is larger, it means that the reference pixel and one of the edge pixels are more likely to be a pair of vascular wall pixels.

[0031] In one embodiment of the present invention, the preset first neighborhood is set to 50 edge pixels on both sides of the same edge as the reference pixel. It should be noted that in other embodiments of the present invention, the preset first neighborhood can be set arbitrarily, which is not limited here.

[0032] Preferably, in one embodiment of the present invention, the method for acquiring blood vessel wall pixels includes: taking all boundary pixels in the cardiovascular image whose blood vessel wall probability is not less than a preset first threshold as blood vessel wall pixels. In one embodiment of the present invention, the preset first threshold is set to 0.85. It should be noted that in other embodiments of the present invention, the preset first threshold can be set voluntarily and is not limited here.

[0033] Since the abnormal circulation of coronary arteries in the cardiovascular system is mostly caused by the accumulation of impurities, which usually does not cause actual deformation, it is necessary to analyze the proximity between the reference vascular wall pixel and other pixel points, that is, in one embodiment of the present invention, the proximity of the vascular wall of each other pixel point in the preset second neighborhood is obtained according to the grayscale difference and gradient difference between the reference vascular wall pixel point and each other pixel point in the preset second neighborhood. Since the vascular walls on both sides of the abnormal circulation area will not be affected and still have strong connectivity in morphology, in the embodiment of the present invention, the reference vascular wall pixel point is extended according to the vascular wall proximity to obtain all extended vascular wall pixel points of the reference vascular wall pixel point; according to the change of the extension direction of each extended vascular wall pixel point, the extension direction mutation index of each extended vascular wall pixel point is obtained; according to the extension direction mutation index, the direction mutation point in the extended vascular wall pixel point is obtained, and the direction mutation point is used to analyze which pixels belong to the abnormal circulation area.

[0034] Preferably, in one embodiment of the present invention, the method for obtaining the blood vessel wall proximity comprises: obtaining the blood vessel wall proximity according to a blood vessel wall proximity calculation formula, the blood vessel wall proximity calculation formula is as follows: In the formula, Indicates the reference blood vessel wall pixel point preset in the second neighborhood The proximity of other pixels to the blood vessel wall; Represents the gray value of the reference blood vessel wall pixel; Indicates the reference blood vessel wall pixel point preset in the second neighborhood Grayscale values ​​of other pixels; Represents the gradient value of the reference blood vessel wall pixel; Indicates the reference blood vessel wall pixel point preset in the second neighborhood The gradient values ​​of other pixels; Represents the gradient direction angle of the reference blood vessel wall pixel; Indicates the reference blood vessel wall pixel point preset in the second neighborhood The gradient direction angles of other pixels.

[0035] In the calculation formula for the degree of proximity of the vascular wall, if the grayscale difference between the reference vascular wall pixel and each other pixel in the preset second neighborhood, the smaller the gradient value difference and the gradient direction difference, the more similar the other pixels are to the reference vascular wall pixel, the closer the other pixels are to the reference vascular wall pixel, that is, the greater the degree of proximity of the other pixels to the vascular wall.

[0036] Preferably, in one embodiment of the present invention, the method for obtaining extended blood vessel wall pixel points includes: selecting a pixel point with the greatest degree of proximity to the blood vessel wall within a preset second neighborhood of a reference blood vessel wall pixel point as a first extended blood vessel wall pixel point of the reference blood vessel wall pixel point, and taking the direction in which the reference blood vessel wall pixel point points to the first extended blood vessel wall pixel point as the extension direction of the reference blood vessel wall pixel point, and taking the angle between the extension direction and the horizontal direction as the extension direction angle of the reference blood vessel wall pixel point.

[0037] The first extended blood vessel wall pixel point is used as a starting point for extension until a preset number of extended blood vessel wall pixel points are obtained, and an extension direction angle of each extended blood vessel wall pixel point is obtained.

[0038] In one embodiment of the present invention, the preset number is set to 10. It should be noted that in other embodiments of the present invention, the preset number can be set arbitrarily and is not limited here.

[0039] Preferably, in one embodiment of the present invention, the method for obtaining the extension direction mutation index includes: sorting a preset number of extended blood vessel wall pixel points according to the extension order to obtain an extended blood vessel wall pixel point sequence; obtaining the extension direction mutation index according to an extension direction mutation index calculation formula, and the extension direction mutation index calculation formula is as follows: In the formula, Indicates the extension direction mutation index of the extended blood vessel wall pixel point; represents the preset number of pixels of the extended blood vessel wall; Indicates the extension direction angle of the last extended blood vessel wall pixel point in the extended blood vessel wall pixel point sequence; Indicates the first pixel in the extended blood vessel wall sequence The extension direction angle of the extended blood vessel wall pixel.

[0040] In the calculation formula of the extension direction mutation index, the first The greater the difference in the extension direction angle between the first extended vascular wall pixel point and the last extended vascular wall pixel point, the greater the mutation in the extension direction. The serial number of the extended vascular wall pixel point is used as the weight of the extension direction angle difference. The larger the serial number, the greater the mutation. The closer the distance between the first extended blood vessel wall pixel and the last extended blood vessel wall pixel, the greater the weight, indicating that the The greater the mutation degree of the extension direction angle between the first extended vascular wall pixel point and the last extended vascular wall pixel point, that is, the greater the extension direction mutation index of the extended vascular wall pixel point.

[0041] Since the shape of the blood vessel wall is relatively smooth, if there is a sudden change in the extension direction during the extension of the blood vessel wall pixels, it usually corresponds to the bend, bifurcation or abnormal flow of the blood vessel. At this time, continue to extend in the direction before the mutation. If you can match the blood vessel wall pixels similar to the direction before the mutation, and pass through the grayscale uneven area suspected of abnormal blood circulation, it means that the extension change is caused by the abnormal circulation area. Therefore, according to the extension direction mutation index of the extended blood vessel wall pixels, the extended blood vessel wall pixels are screened to obtain the direction mutation points, and then analyze which pixels belong to the abnormal circulation area.

[0042] Preferably, in one embodiment of the present invention, the method for obtaining the direction mutation point includes: obtaining a preset second threshold, and when the ratio between the extension direction mutation index of the extended blood vessel wall pixel point and the previous extended blood vessel wall pixel point is not less than the preset second threshold, taking the extended blood vessel wall pixel point as the direction mutation point.

[0043] In one embodiment of the present invention, the preset second threshold is set to 1.5. It should be noted that in other embodiments of the present invention, the preset second threshold can be set arbitrarily, and is not limited here.

[0044] The abnormal circulation area acquisition module 103 is used to select a pixel point as a target pixel point in the preset moving window of each directional mutation point, and obtain the possible abnormality of the target pixel point according to the proximity of the target pixel point to the vascular wall, position distribution, gradient distribution and grayscale distribution between the target pixel point and the window center in the moving window; according to the possible abnormality of all pixels, the abnormal circulation area in the cardiovascular image is obtained.

[0045] Since the blood vessel walls on both sides of the abnormal flow area will not be affected and still have strong connectivity in morphology, when the reference blood vessel wall pixel continues to extend in the direction before the change, if it matches a blood vessel wall with strong connectivity within a certain distance after the extension, it means that the change may be caused by the abnormal flow area in the blood vessel. Since all directional mutation points have been obtained in the above steps, in the embodiment of the present invention, any pixel point in the preset moving window of each directional mutation point is selected as the target pixel point, and the possible degree of abnormality of the target pixel point is obtained according to the proximity, position distribution, gradient distribution and grayscale distribution of the blood vessel wall between the target pixel point and the window center in the moving window, and finally the abnormal flow area in the cardiovascular image is screened out.

[0046] Preferably, in one embodiment of the present invention, a method for obtaining the possible degree of abnormality of a target pixel point includes: obtaining the degree of matching between the extended vascular wall and the real vascular wall according to the degree of proximity and distance distribution between each directional mutation point and each pixel point in a preset moving window. In one embodiment of the present invention, a method for obtaining the degree of matching is provided, which specifically includes: establishing a directional mutation point as the center, The rectangular window is used as the preset moving window, and the rectangular window is moved in the extension direction of the previous extended blood vessel wall pixel point of the direction mutation point, and the moving step is set to 200. The trajectory of the window center during the moving process is recorded as the moving trajectory. Among them, the parameters of the rectangular window and the moving step can be set by yourself and are not limited here.

[0047] The matching degree is obtained according to the matching degree calculation formula. The matching degree calculation formula is as follows: In the formula, Indicates the matching degree between the extended vessel wall and the real vessel wall; Indicates the number of pixels in the rectangular window during the moving process; The previous extended blood vessel wall pixel representing the direction mutation point and the first pixel in the rectangular window moving process The proximity of the blood vessel wall between pixels; Indicates the preset third neighborhood The distance between the pixel point and the moving trajectory.

[0048] In the matching degree calculation formula, the previous extended blood vessel wall pixel point of the direction mutation point and the first pixel point within the rectangular window moving process are The closer the blood vessel walls are to each other, the The more likely the pixel is a blood vessel wall pixel, that is, the greater the matching degree between the extended blood vessel wall and the real blood vessel wall, at this time, if the The smaller the difference between the distance between the pixel point and the moving trajectory, the higher the matching degree between the extended blood vessel wall and the real blood vessel wall.

[0049] The extended blood vessel wall pixel points are obtained according to the calculation formula of the possible degree of abnormality. The calculation formula of the possible degree of abnormality is as follows: In the formula, Indicates the possible degree of abnormality of the target pixel; Indicates the matching degree between the extended vessel wall and the real vessel wall; Indicates the number of pixels within the preset third neighborhood of the target pixel; Indicates the third neighborhood of the target pixel. The gradient value of each pixel; Indicates the third neighborhood of the target pixel. The gray value of each pixel; Represents the standard deviation function.

[0050] In the calculation formula of the abnormal possibility, the higher the matching degree between the extended blood vessel wall and the real blood vessel wall, the more likely the blood vessel wall is blocked. At this time, the target pixel is more likely to be in the abnormal flow area. The overall gradient value of the third neighborhood of the target pixel is preset. The larger it is, the higher the intensity of the grayscale change around the target pixel, which means the more likely the target pixel is in an abnormal circulation area; the larger the grayscale standard deviation of the preset third neighborhood of the target pixel, the more uneven the grayscale distribution around the target pixel, which means the more likely the target pixel is in an abnormal circulation area, that is, the more likely the target pixel is abnormal.

[0051] In one embodiment of the present invention, the degree of abnormal possibility of all pixels in the moving window is obtained, a preset third threshold is set, all pixels whose degree of abnormal possibility is not less than the preset third threshold are regarded as pixels in the abnormal circulation area, and the positions of all pixels in the abnormal circulation area are combined as the abnormal circulation area. The preset third threshold is set to 0.65. It should be noted that in other embodiments of the present invention, the preset third threshold can be set voluntarily and is not limited here.

[0052] The navigation and positioning module 104 is used to perform image enhancement on the abnormal circulation area to obtain a final cardiovascular image; and perform surgical navigation and positioning according to the final cardiovascular image.

[0053] After the abnormal circulation area is determined, the digital subtraction angiography image is locally enhanced. Since too high an image enhancement intensity may distort the pixels in the abnormal circulation area, certain restrictions need to be imposed when enhancing the pixels in the abnormal circulation area. Therefore, the target grayscale mapping range of the abnormal circulation area is analyzed. In the embodiment of the present invention, the target grayscale mapping range calculation formula is as follows: In the formula, The target grayscale mapping range representing the abnormal circulation area; Indicates the possible degree of abnormality of all pixels in the abnormal circulation area; Indicates the maximum grayscale mapping range of the abnormal circulation area, which is the difference between the maximum grayscale and the minimum grayscale, which is 255.

[0054] The patient's cardiovascular image is enhanced using linear contrast stretching technology to obtain a final cardiovascular image. It should be noted that the linear contrast stretching technology is a technical means well known to those skilled in the art and will not be described in detail here.

[0055] The abnormal circulation area is marked using the connected domain labeling algorithm, and the abnormal circulation area on the patient's cardiovascular system is finally displayed on the display device through visualization technology, providing decision support for relevant personnel.

[0056] At this point, real-time image navigation and positioning in cardiovascular interventional surgery are completed.

[0057] In summary, a cardiovascular image of each patient is obtained. In the cardiovascular image, an edge pixel point is selected as a reference pixel point, and the possibility of the reference pixel point being a vascular wall pixel point is obtained according to the grayscale difference and gradient difference between the reference pixel point and each other pixel point in the preset first neighborhood; all pixels are screened according to the vascular wall possibility to obtain all vascular wall pixels in the cardiovascular image; one vascular wall pixel point is selected as a reference vascular wall pixel point; according to the grayscale difference and gradient difference between the reference vascular wall pixel point and each other pixel point in the preset second neighborhood, the vascular wall proximity of each other pixel point in the preset second neighborhood is obtained; according to the vascular wall proximity, the reference vascular wall pixel point is extended to obtain all extended vascular wall pixels of the reference vascular wall pixel point; according to the extension direction change of each extended vascular wall pixel point, the extension direction mutation index of each extended vascular wall pixel point is obtained; according to the extension direction mutation index, the direction mutation point in the extended vascular wall pixel point is obtained. In the preset moving window of each directional mutation point, a pixel point is selected as the target pixel point. According to the proximity degree, position distribution, gradient distribution and grayscale distribution of the target pixel point, the possible abnormality of the target pixel point is obtained; according to the possible abnormality of all pixels, the abnormal circulation area in the cardiovascular image is obtained. The image of the abnormal circulation area is enhanced to obtain the final cardiovascular image; surgical navigation and positioning are performed based on the final cardiovascular image.

[0058] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0059] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A real-time image navigation and positioning system for cardiovascular intervention surgery, characterized in that: The system comprises: an image acquisition module for acquiring a cardiovascular image of each patient; a mutation point recognition module for selecting an edge pixel point in the cardiovascular image as a reference pixel point, and obtaining the vascular wall possibility that the reference pixel point is a vascular wall pixel point according to the grayscale difference and gradient difference between the reference pixel point and each other pixel point in a preset first neighborhood; screening all pixel points according to the vascular wall possibility to obtain all vascular wall pixel points in the cardiovascular image; selecting an vascular wall pixel point as a reference vascular wall pixel point; obtaining the vascular wall proximity of each other pixel point in the preset second neighborhood according to the grayscale difference and gradient difference between the reference vascular wall pixel point and each other pixel point in the preset second neighborhood; extending the reference vascular wall pixel point according to the vascular wall proximity to obtain a reference vascular wall pixel point. Consider all extended vascular wall pixel points of the vascular wall pixel points; obtain the extension direction mutation index of each extended vascular wall pixel point according to the extension direction change of each extended vascular wall pixel point; obtain the direction mutation point in the extended vascular wall pixel points according to the extension direction mutation index; an abnormal circulation area acquisition module is used to select a pixel point as a target pixel point in a preset moving window of each direction mutation point, and obtain the abnormal possibility of the target pixel point according to the vascular wall proximity, position distribution, gradient distribution and grayscale distribution of the target pixel point; obtain the abnormal circulation area in the cardiovascular image according to the abnormal possibility of all pixels; a navigation and positioning module is used to perform image enhancement on the abnormal circulation area to obtain a final cardiovascular image; and perform surgical navigation and positioning according to the final cardiovascular image.

2. A real-time image navigation and positioning system for cardiovascular interventional surgery according to claim 1, characterized in that: The method for obtaining the possibility of the blood vessel wall includes: obtaining the possibility that the reference pixel and each other edge pixel are paired blood vessel wall pixels according to the grayscale difference and gradient direction difference between the reference pixel and other edge pixels; obtaining the possibility of the blood vessel wall according to the blood vessel wall possibility calculation formula, and the blood vessel wall possibility calculation formula is as follows: In the formula, Represents the possibility of blood vessel wall at the reference pixel; Represents the gradient value of the reference pixel; Indicates the maximum gradient value of edge pixels in cardiovascular images; Indicates the reference pixel and the The possibility that all edge pixels are blood vessel wall pixels; Indicates the number of pixels other than the reference pixel in the preset first neighborhood; represents the normalization function; represents the maximum value function; Represents the union function.

3. A real-time image navigation and positioning system for cardiovascular interventional surgery according to claim 2, characterized in that: The method for obtaining the paired possibility that the reference pixel point and each other edge pixel point are all blood vessel wall pixels includes: establishing a Cartesian coordinate system for the cardiovascular image, taking the angle between the gradient direction of the edge pixel point and the positive direction of the x-axis as the gradient direction angle; obtaining the paired possibility according to a paired possibility calculation formula, and the paired possibility calculation formula is as follows: In the formula, Indicates the reference pixel and the The possibility that all edge pixels are blood vessel wall pixels; Indicates the reference pixel and the The gray value of the pixel with the smaller gray value among the edge pixels; Indicates the reference pixel and the The gray value of the pixel with the larger gray value among the edge pixels; Indicates the reference pixel and the The gradient direction angle of the pixel with the larger gradient direction angle among the edge pixels; Indicates the reference pixel and the The gradient direction angle of the pixel with the smaller gradient direction angle among the edge pixels; represents the absolute value function.

4. A real-time image navigation and positioning system for cardiovascular interventional surgery according to claim 1, characterized in that: The method for acquiring blood vessel wall pixel points includes: taking all boundary pixel points in the cardiovascular image whose blood vessel wall probability is not less than a preset first threshold as blood vessel wall pixel points.

5. The real-time image navigation and positioning system in cardiovascular intervention surgery according to claim 1, characterized in that: The method for obtaining the blood vessel wall proximity degree includes: obtaining the blood vessel wall proximity degree according to a blood vessel wall proximity degree calculation formula, and the blood vessel wall proximity degree calculation formula is as follows: In the formula, Indicates the reference blood vessel wall pixel point preset in the second neighborhood The proximity of other pixels to the blood vessel wall; Represents the gray value of the reference blood vessel wall pixel; Indicates the reference blood vessel wall pixel point preset in the second neighborhood Grayscale values ​​of other pixels; Represents the gradient value of the reference blood vessel wall pixel; Indicates the reference blood vessel wall pixel point preset in the second neighborhood The gradient values ​​of other pixels; Represents the gradient direction angle of the reference blood vessel wall pixel; Indicates the reference blood vessel wall pixel point preset in the second neighborhood The gradient direction angles of other pixels.

6. A real-time image navigation and positioning system for cardiovascular interventional surgery according to claim 1, characterized in that: The method for acquiring the pixel points of the extended blood vessel wall comprises: A pixel point with the greatest degree of proximity to the blood vessel wall is selected within a preset second neighborhood of a reference blood vessel wall pixel point as a first extended blood vessel wall pixel point of the reference blood vessel wall pixel point, and a direction in which the reference blood vessel wall pixel point points to the first extended blood vessel wall pixel point is used as an extension direction of the reference blood vessel wall pixel point, and an angle between the extension direction and the horizontal direction is used as an extension direction angle of the reference blood vessel wall pixel point; the first extended blood vessel wall pixel point is used as a starting point for extension until a preset number of extended blood vessel wall pixel points are obtained, and an extension direction angle of each extended blood vessel wall pixel point is obtained.

7. A real-time image navigation and positioning system for cardiovascular interventional surgery according to claim 1, characterized in that: The method for obtaining the extension direction mutation index includes: sorting a preset number of extended blood vessel wall pixel points according to the extension order to obtain an extended blood vessel wall pixel point sequence; obtaining the extension direction mutation index according to an extension direction mutation index calculation formula, and the extension direction mutation index calculation formula is as follows: In the formula, Indicates the extension direction mutation index of the extended blood vessel wall pixel point; Represents the preset number of pixels of the extended blood vessel wall; Indicates the extension direction angle of the last extended blood vessel wall pixel point in the extended blood vessel wall pixel point sequence; Indicates the first pixel in the extended blood vessel wall sequence The extension direction angle of the extended blood vessel wall pixel.

8. The real-time image navigation and positioning system in cardiovascular intervention surgery according to claim 1, characterized in that: The method for acquiring the direction mutation point includes: obtaining a preset second threshold, and when the ratio between the extension direction mutation index of the extended blood vessel wall pixel point and the previous extended blood vessel wall pixel point is not less than the preset second threshold, taking the extended blood vessel wall pixel point as the direction mutation point.

9. The real-time image navigation and positioning system in cardiovascular intervention surgery according to claim 1, characterized in that: The method for obtaining the abnormal possibility degree includes: obtaining the matching degree between the extended blood vessel wall and the real blood vessel wall according to the blood vessel wall proximity degree and distance distribution between each directional mutation point and each pixel point in the preset moving window; obtaining the extended blood vessel wall pixel point according to the abnormal possibility degree calculation formula, and the abnormal possibility degree calculation formula is as follows: In the formula, Indicates the possible degree of abnormality of the target pixel; Indicates the matching degree between the extended vessel wall and the real vessel wall; Indicates the number of pixels within the preset third neighborhood of the target pixel; Indicates the third neighborhood of the target pixel. The gradient value of each pixel; Indicates the third neighborhood of the target pixel. The gray value of each pixel; Represents the standard deviation function.

10. A real-time image navigation and positioning system for cardiovascular interventional surgery according to claim 9, characterized in that: The method for obtaining the degree of matching between the extended blood vessel wall and the real blood vessel wall includes: establishing a rectangular window with the direction mutation point as the center, moving the rectangular window in the extension direction of the previous extended blood vessel wall pixel point of the direction mutation point, and recording the trajectory of the window center during the movement as the movement trajectory; obtaining the matching degree according to the matching degree calculation formula, and the matching degree calculation formula is as follows: In the formula, Indicates the matching degree between the extended vessel wall and the real vessel wall; Indicates the number of pixels in the rectangular window during the moving process; The previous extended blood vessel wall pixel representing the direction mutation point and the first pixel in the rectangular window moving process The proximity of the blood vessel wall between pixels; Indicates the preset third neighborhood The distance between the pixel point and the moving trajectory.

Images