Computer-implemented method for determining tubular stenosis rate and related products

Abstract

The present invention relates to a computer-implemented method, device, apparatus, computer-readable storage medium, and device for predicting the risk of cognitive dysfunction for determining the stenosis rate of a tubular object. The method comprises: in response to receiving a straightened image of a tubular object, obtaining a cross-sectional image at each center point of the straightened tubular object in the straightened image; detecting the cross-sectional radius or cross-sectional diameter of the tubular cross section in the cross-sectional image at each center point; determining the average radius or average diameter of the straightened tubular object based on the weighted average of the cross-sectional radius or cross-sectional diameter at each center point; and determining the stenosis rate of the tubular object based on the minimum cross-sectional radius among each cross-sectional radius and the average radius, or based on the minimum cross-sectional diameter among each cross-sectional diameter and the average diameter. The method according to an embodiment of the present invention has the characteristics of fast detection speed, low average error, and high accuracy.

Description

Technical Field

[0001] The present invention generally relates to the field of image processing technology. More specifically, the present invention relates to a computer-implemented method, device, apparatus, computer-readable storage medium for determining tubular stenosis rate, and a device for predicting the risk of cognitive dysfunction. Background Art

[0002] At present, the analysis of stenosis rate of tubular structure tissue (such as blood vessels, etc.) mainly relies on manual measurement by medical personnel, but manual measurement is not ideal in terms of efficiency and accuracy. In addition, the current determination of stenosis rate mainly relies on the measurement and analysis of two-dimensional tubular images, which cannot accurately detect the stenosis rate of tubular objects with irregular cross-sections. Therefore, a method that can quickly and accurately determine the stenosis rate of tubular objects is urgently needed. Summary of the invention

[0003] In view of the technical problems mentioned above, the technical solutions of the present invention provide, in multiple aspects, a computer-implemented method, device, apparatus, computer-readable storage medium for determining tubular stenosis rate, and a device for predicting the risk of cognitive dysfunction.

[0004] In a first aspect of the present invention, a computer-implemented method for determining the stenosis rate of a tubular object is provided, comprising: in response to receiving a straightened image of a tubular object, acquiring a cross-sectional image at each center point of the straightened tubular object in the straightened image; detecting the cross-sectional radius or cross-sectional diameter of the tubular cross section in the cross-sectional image at each center point; determining the average radius or average diameter of the straightened tubular object based on a weighted average of the cross-sectional radii or cross-sectional diameters at each center point; and determining the stenosis rate of the tubular object based on the minimum cross-sectional radius among each cross-sectional radii and the average radius, or based on the minimum cross-sectional diameter among each cross-sectional diameter and the average diameter.

[0005] In one embodiment, detecting the cross-sectional radius or cross-sectional diameter at each center point includes: identifying the center point of the tubular cross section in each cross-sectional image, and setting a plurality of cross-sectional lines on the cross-sectional image with the center point as the intersection point; detecting the intersection of each cross-sectional line with the edge of the tubular object in the cross-sectional image to determine the sampling point in the cross-sectional image; and determining the cross-sectional radius or cross-sectional diameter at the center point based on the average value of the sampling radius between each sampling point and the intersection point.

[0006] In another embodiment, determining the sampling point in the cross-sectional image includes: dividing each intersection line into two intersection sub-lines at the intersection point; and determining the intersection point on each intersection sub-line that is closest to the intersection point as the sampling point.

[0007] In another embodiment, detecting the cross-sectional radius at each center point includes: identifying the center point of the tubular cross section in each cross-sectional image, and setting a plurality of radial lines toward the edge of the tubular with the center point as a starting point; detecting the intersection of each radial line with the edge of the tubular in the cross-sectional image to determine the sampling point in the cross-sectional image; and determining the cross-sectional radius at the center point based on the average value of the sampling radius between each sampling point and the center point.

[0008] In one embodiment, the method further includes: calculating the relative difference between each sampling radius and the cross-sectional radius of the cross-sectional image in which it is located; in response to the relative difference being greater than a preset threshold, determining that the sampling radius having a relative difference greater than the preset threshold is an abnormal sampling radius; and updating the cross-sectional radius based on the average value of other sampling radii except the abnormal sampling radius.

[0009] In another embodiment, the method further includes: determining a first length at the abnormal sampling radius based on a difference between a maximum sampling radius among one or more adjacent abnormal sampling radii and an updated cross-sectional radius; determining a first width at the abnormal sampling radius based on a width of an area where the maximum sampling radius exceeds the updated cross-sectional radius; and determining that an abnormal protrusion exists at the abnormal sampling radius in response to the first length being greater than a first length threshold and the first width being greater than a first width threshold.

[0010] In another embodiment, determining the average radius or the average diameter includes: performing a weighted average operation on each cross-sectional radii based on the distance between the center point of the minimum cross-sectional radius among each cross-sectional radii in the straightened image and each center point to determine the average radius; or performing a weighted average operation on each cross-sectional diameter based on the distance between the center point of the minimum cross-sectional diameter among each cross-sectional diameter in the straightened image and each center point to determine the average diameter.

[0011] In one embodiment, performing a weighted average operation includes: constructing a function of the distance according to a preset rule; determining a weight value of each cross-sectional radius or each cross-sectional diameter based on the function; and performing a weighted average operation on each cross-sectional radius or each cross-sectional diameter according to the weight value.

[0012] In another embodiment, the function includes at least one of a linear function, a quadratic function, an inverse proportional function, and a Gaussian function.

[0013] In yet another embodiment, the tubular object comprises a cerebral arterial vessel.

[0014] In a second aspect of the present invention, there is provided an apparatus for determining the stenosis rate of a tubular object, comprising: a cross-sectional image acquisition module for acquiring, in response to receiving a straightened image of the tubular object, a cross-sectional image at each center point of the straightened tubular object in the straightened image; a detection module for detecting the cross-sectional radius or cross-sectional diameter of the tubular cross section in the cross-sectional image at each center point; a mean value determination module for determining the average radius or average diameter of the straightened tubular object based on a weighted average of the cross-sectional radii or cross-sectional diameters at each center point; and a stenosis rate determination module for determining the stenosis rate of the tubular object based on the minimum cross-sectional radius among the cross-sectional radii and the average radius, or based on the minimum cross-sectional diameter among the cross-sectional diameters and the average diameter.

[0015] In a third aspect of the present invention, there is provided an apparatus for determining a stenosis rate of a tubular object, comprising at least one processor; and a memory storing program instructions, wherein when the program instructions are executed by the at least one processor, the apparatus executes the method according to any one of the first aspect of the present invention.

[0016] In a fourth aspect of the present invention, a computer-readable storage medium is provided, which stores a computer-implemented program for determining a tubular stenosis rate, and when the program is executed by a processor, the method according to any one of the first aspects of the present invention is executed.

[0017] In the fifth aspect of the present invention, a device for predicting the risk of cognitive dysfunction is provided, comprising: a stenosis rate determination module, used to determine the stenosis rate of each cerebral artery according to any method described in the first aspect of the present invention; a binarization processing module, used to binarize the stenosis rates of multiple cerebral arteries; a dimensionality reduction module, used to perform feature dimensionality reduction processing on the processing results of the binarization processing; and a classification model, used to classify the dimensionality reduction results after the feature dimensionality reduction processing, so as to predict the risk of cognitive dysfunction in the owners of the multiple cerebral arteries.

[0018] Based on the above description of the scheme of the present invention, those skilled in the art can understand the scheme described in the above embodiment, and can realize the automatic detection of the machine by detecting the cross-sectional radius or cross-sectional diameter of the cross-sectional image at each center point in the straightened image through computer implementation, which is conducive to improving the detection speed; and by acquiring the cross-sectional image to obtain the three-dimensional image information at each center point of the straightened image, it is conducive to realizing the effective analysis and detection of the cross section of any shape of tubular object. By determining the stenosis rate of the tubular object based on the weighted average radius or average diameter, it is possible to fully consider the different effects of each cross-sectional radius or each cross-sectional diameter of the tubular object on the average radius or average diameter, which is conducive to further improving the accuracy of the detection result of the stenosis rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0020] Figure 1 is a flowchart showing a method for determining a tubular object stenosis rate implemented by a computer according to an embodiment of the present invention;

[0021] Figure 2 is a schematic diagram showing the effect of obtaining a straightened image according to an embodiment of the present invention;

[0022] Figure 3 Schematic diagram showing a process of removing burrs on a skeleton line according to an embodiment of the present invention, wherein Figure 3 Figure (a) is a schematic diagram showing a straightened tubular object with burrs on the skeleton line according to an embodiment of the present invention. Figure 3 Figure (b) is a schematic diagram showing the center line after burrs are removed according to an embodiment of the present invention;

[0023] Figure 4 is a schematic diagram showing a process of removing the ring structure on the skeleton line according to an embodiment of the present invention, wherein Figure 4 Figure (a) is a schematic diagram showing a straightened tubular object with a ring structure on the skeleton line according to an embodiment of the present invention, Figure 4 Figure (b) is a schematic diagram showing the center line after the annular structure is removed according to an embodiment of the present invention;

[0024] Figure 5 is a schematic diagram showing a process of connecting skeleton wires in the same straightened tubular object according to an embodiment of the present invention, wherein Figure 5 Figure (a) is a schematic diagram showing that the skeleton line of the same straightened tubular object according to an embodiment of the present invention includes multiple nodes. Figure 5 Figure (b) is a schematic diagram showing the center line connected to the skeleton line of the same straightened tubular object according to an embodiment of the present invention;

[0025] Figure 6 is a flow chart showing a method for detecting a cross-sectional radius according to an embodiment of the present invention;

[0026] Figure 7 is a flow chart showing a method for determining an abnormal protrusion according to an embodiment of the present invention;

[0027] Figure 8 is a schematic cross-sectional view showing the presence of an abnormal protrusion according to an embodiment of the present invention;

[0028] Fig. 9is a schematic diagram showing the arrangement of cross lines according to an embodiment of the present invention, wherein Fig. 9 FIG. (a) is a schematic diagram showing the arrangement of cross lines for a circular blood vessel cross section according to an embodiment of the present invention. Fig. 9 FIG. (b) is a schematic diagram showing the setting of cross lines for an elliptical blood vessel cross section according to an embodiment of the present invention. Fig. 9 FIG. (c) is a schematic diagram showing a method of setting a cross line for a concave blood vessel cross section according to an embodiment of the present invention. Fig. 9 FIG. (d) is a schematic diagram showing setting of cross lines for a blood vessel cross section having blood vessel adhesion or non-blood vessel area according to an embodiment of the present invention. Fig. 9 Figure (e) is a schematic diagram showing the arrangement of cross lines for a blood vessel cross section having protrusions according to an embodiment of the present invention;

[0029] Fig.10 is a comparison chart showing various cross-sectional radius measurement methods;

[0030] Fig.11 is a schematic diagram showing the principle of determining an average radius according to an embodiment of the present invention;

[0031] Fig.12 is a schematic diagram showing the relationship between the weight value and the distance according to an embodiment of the present invention;

[0032] Fig.13 is a graph showing the error comparison between various stenosis rate determination methods and the radiologist's standard results; and

[0033] Fig.14 FIG. 4 is a schematic diagram of a system for determining a stenosis rate of a tubular object according to an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present disclosure.

[0035] It should be understood that the terms "first", "second", "third" and "fourth" etc. in the claims, specifications and drawings of the present invention are used to distinguish different objects rather than to describe a specific order. The terms "include" and "comprise" used in the specification and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their collections.

[0036] It should also be understood that the terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit the present disclosure. As used in the specification and claims of the present invention, the singular forms of "a", "an" and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" used in the specification and claims of the present invention refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations.

[0037] As used in this specification and claims, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" may be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0038] The present invention provides a new and feasible solution to the deficiencies of the prior art. In particular, the method of the embodiment of the present invention can obtain cross-sectional images at each center point in the straightened image, detect the cross-sectional radius or cross-sectional diameter of the tubular object based on each cross-sectional image, and determine the stenosis rate of the tubular object based on the weighted average radius or average diameter, which can not only realize automatic real-time detection, but also help improve detection accuracy, providing strong technical support for clinical applications and scientific research.

[0039] Through the following description, those skilled in the art can understand that the present invention also provides implementation methods in multiple embodiments that are conducive to further improving the stenosis rate detection effect. For example, in some embodiments, multiple radial lines can be set on the cross-sectional image so that the method according to the embodiment of the present invention can be applicable to the cross-sectional radius detection of tubular cross-sections of various shapes. In other embodiments, the cross-sectional radius can be updated by detecting whether there is an abnormal sampling radius in the sampling radius and eliminating the influence of the abnormal sampling radius, thereby facilitating the improvement of the detection accuracy of the cross-sectional radius. The specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0040] Figure 1 FIG. 4 is a flowchart showing a method for determining a tubular object stenosis rate according to an embodiment of the present invention implemented by a computer. Figure 1As shown in , method 100 may include: in step 110, in response to receiving the straightened image of the tubular object, obtaining a cross-sectional image at each center point of the straightened tubular object in the straightened image. In some embodiments, the tubular object may include at least one of tubular tissues such as blood vessels, small intestines, and fallopian tubes. In one embodiment of the present invention, the tubular object may include at least one of cardiovascular, cerebral, capillary, arterial, and venous blood vessels. In another embodiment, the tubular object may include a cerebral arterial blood vessel.

[0041] In some embodiments, the straightened image may be an image obtained by straightening an image containing a tubular object. Figure 2 As shown in , a straightened image can be obtained by straightening the tubular object 201, wherein the straightened image may include a straightened tubular object 202 after the tubular object 201 is straightened. In some embodiments, the straightening process may include performing a translation transformation and / or a rotation transformation on the coordinates of the tubular object in the original image. In other embodiments, the straightened image is a three-dimensional image.

[0042] In some embodiments, the center line of the straightened tubular object may be a refinement of the entire straightened tubular object, and may be used to reflect the overall skeleton structure of the straightened tubular object. In other embodiments, the center line of the straightened tubular object may be the skeleton line of the straightened tubular object. In still other embodiments, each point on the center line may be the center point of the cross section of the straightened tubular object, and the center line may be a collection of the center points. In some application scenarios, for example, when the straightened tubular object has an irregularly shaped cross section, the center point of the cross section on the center line of the straightened tubular object may not be the center of the circle in the strict sense. In some embodiments, obtaining a cross-sectional image may be achieved by obtaining the grayscale value of each coordinate on the cross section at each center point. In other embodiments, obtaining a cross-sectional image may be achieved by, for example, cutting the straightened tubular object at each center point to intercept the cross-sectional image at each center point.

[0043] In one embodiment, step 110 may include: performing skeleton extraction on the straightened tubular object in the straightened image to determine the center line of the straightened tubular object. The straightened tubular object in the straightened image may be thinned using skeleton extraction technology (or binary image thinning technology) to obtain the skeleton line of the straightened tubular object. In some embodiments, the skeleton line may be used as the center line. Each straightened tubular object may correspond to a center line. In some embodiments, the straightened image includes a straightened tubular object, and skeleton extraction may be performed on the straightened tubular object to determine the center line of the straightened tubular object. In other embodiments, the straightened image may include multiple straightened tubular objects, and skeleton extraction may be performed on all or part of the straightened tubular objects in the straightened image to determine the center line of all or part of the straightened tubular objects.

[0044] Next, in step 120, the cross-sectional radius or cross-sectional diameter of the tubular cross section in the cross-sectional image at each center point may be detected. In some embodiments, the cross-sectional radius may be obtained by detecting the cross-sectional image by methods such as the minimum distance method, the area measurement method, etc. In another embodiment, the cross-sectional diameter may be determined based on twice the cross-sectional radius.

[0045] Then, the process can proceed to step 130, and the average radius or average diameter of the straightened tubular object can be determined based on the weighted average of the cross-sectional radius or cross-sectional diameter at each center point. In some embodiments, the weight value of each cross-sectional radius or each cross-sectional diameter can be determined according to the contribution of the cross-sectional radius or cross-sectional diameter at each center point to the overall shape of the tubular object, and the weighted average operation of each cross-sectional radius or each cross-sectional diameter can be performed according to the weight value of each cross-sectional radius or each cross-sectional diameter to obtain the average radius or average diameter of the straightened tubular object. According to such a setting, the influence of each cross-sectional radius or cross-sectional diameter on the overall shape evaluation of the tubular object can be taken into account, so that the measurement result of the average radius or average diameter that is closer to the actual state of the tubular object can be obtained. In other embodiments, the average diameter can be determined based on twice the average radius.

[0046] Further, in step 140, the stenosis rate of the tubular object may be determined based on the minimum cross-sectional radius and the average radius among the cross-sectional radii, or based on the minimum cross-sectional diameter and the average diameter among the cross-sectional diameters. The minimum cross-sectional radius may be the minimum value among the cross-sectional radii of the straightened tubular object. The minimum cross-sectional diameter may be the minimum value among the cross-sectional diameters of the straightened tubular object.

[0047] In one embodiment, the stenosis rate of the tubular object can be determined based on the ratio of the minimum cross-sectional radius to the average radius among the cross-sectional radii, or based on the ratio of the minimum cross-sectional diameter to the average diameter among the cross-sectional diameters. In another embodiment, the stenosis rate of the tubular object can be calculated based on the following formula:

[0048]

[0049] Among them, s i represents the stenosis rate, Indicates the minimum cross-sectional radius among all cross-sectional radii. represents the mean radius.

[0050] Combination of the above Figure 1 and Figure 2The method for determining the stenosis rate of a tubular object according to an embodiment of the present invention is described exemplarily. It is understandable that the method according to an embodiment of the present invention may not be limited to the above steps. For example, in one embodiment, step 110 may also include: preprocessing the skeleton line obtained by skeleton extraction to generate a preprocessed center line, wherein the preprocessing may include at least one of the following: removing burrs on the skeleton line; removing annular structures on the skeleton line; and connecting the skeleton lines in the same straightened tubular object. In some embodiments, the skeleton line may include a collection of multiple nodes (or center points). For ease of understanding, the following will be combined with Figure 3-Figure 5 An exemplary description is given.

[0051] Figure 3 FIG. 2 is a schematic diagram showing a process of removing burrs on a skeleton line according to an embodiment of the present invention. Figure 3 As shown in FIG. (a), skeleton extraction of the straightened tubular object 301 can obtain a skeleton line 302, and burrs 303 are connected at the nodes 304 of the skeleton line 302. In some application scenarios, for a complete straightened tubular object, the node set on its skeleton line usually includes a starting point and an ending point, and all nodes including the starting point and the ending point should theoretically appear only once. However, for the skeleton line 302 with burrs, the node 304 appears once in the node set as the starting point or the ending point of the burr 303, and appears once as a non-starting and ending point of the skeleton line 302, that is, in the node set of the skeleton line 302, the node 304 appears twice. In other embodiments, the length between the starting and ending points of the burr 303 is shorter.

[0052] Based on the above characteristics of the burr 303, by traversing the skeleton line, the location of the node 304 where the burr 303 appears can be determined based on the number of times the center point (or node) on the skeleton line appears in the node set and / or the length of the skeleton line, so that the burr 303 can be removed to obtain, for example Figure 3 (b) shows the center line 305 after the burrs are removed.

[0053] Figure 4 FIG. 2 is a schematic diagram showing a process of removing a ring structure on a skeleton line according to an embodiment of the present invention. Figure 4As shown in FIG. 4( a), skeleton extraction is performed on the straightened tubular object 401 to obtain a skeleton line 402, wherein a ring structure consisting of a first skeleton line 405 and a second skeleton line 406 exists between a first node 403 and a second node 404 on the skeleton line 402. In some embodiments, in the node set of the skeleton line 402 having the ring structure, the first node 403 and the second node 404 both appear at least twice. Based on the feature of the ring structure, the positions of the first node 403 and the second node 404 can be found by traversing the skeleton line, and the first skeleton line 405 or the second skeleton line 406 in the ring structure can be selectively deleted according to the length relationship between the first skeleton line 405 and the second skeleton line 406, so as to obtain, for example Figure 4 (b) shows the center line 407 after the ring structure is removed.

[0054] In other embodiments, in response to the length ratio of the first skeleton line 405 and the second skeleton line 406 being greater than the second threshold, one of the first skeleton line 405 and the second skeleton line 406 may be randomly removed. In still other embodiments, the second threshold may include 0.8 to 1.2. In some embodiments, in response to the length ratio of the first skeleton line and the second skeleton line being less than the third threshold, a shorter center line in the annular structure may be removed to obtain a center line after the annular structure is removed. In other embodiments, the third threshold may be, for example, 1 / 3. According to such a setting, the influence of errors in adhesion or straightening of the tubular object on the acquisition of cross-sectional images based on each center point can be eliminated.

[0055] In some application scenarios, the distance between two nodes forming a ring structure may not be limited to Figure 4 The first skeleton line and the second skeleton line shown in the figure may also have a larger number of ring structure skeleton lines. In this scenario, the method for removing the ring structure can be combined with the above method. Figure 4 Similar to the description, for example, every two skeleton lines may be compared and removed until only one skeleton line is left.

[0056] Figure 5 Schematic diagram showing the process of connecting skeleton wires in the same straightened tubular object according to an embodiment of the present invention. Figure 5 As shown in FIG. 5( a ), skeleton extraction is performed on the straightened tubular object 501 to obtain a skeleton line 502. In some embodiments, the skeleton line 502 of the same straightened tubular object may include multiple nodes (such as 503, 504, 505, and 506 in the figure), so that the skeleton line 502 is divided into multiple tube segments. In response to the absence of burrs and / or ring structures between the multiple nodes, the skeleton lines between the multiple tube segments can be connected to form a complete skeleton line (such as Figure 5In some other embodiments, the skeleton lines connected within the same straightened tubular object may be connected after the burrs and / or the ring structures are removed.

[0057] Combination of the above Figure 3-Figure 5 The preprocessing of the skeleton line according to the embodiment of the present invention is described exemplarily. It can be understood that through the above preprocessing operation, the skeleton line extracted from the tubular skeleton can be corrected to eliminate the influence of straightening processing errors or tubular adhesions on the acquisition of cross-sectional images, which is conducive to reducing the errors that may be generated when the cross-sectional radius is detected in the subsequent cross-sectional images, and can significantly improve the accuracy of determining the stenosis rate. After obtaining the center line of the straightened tubular object in the straightened image, the cross-sectional images at each center point on the center line can be obtained, and the cross-sectional radius or cross-sectional diameter can be determined based on the cross-sectional image. A specific embodiment of detecting the cross-sectional radius will be described below.

[0058] Figure 6 600 is a flowchart showing a method for detecting a cross-sectional radius according to an embodiment of the present invention. Figure 1 The step 120 described above is a specific embodiment of the present invention. Figure 1 The description of step 120 may also be applied to the following description of method 600 .

[0059] like Figure 6 As shown in , method 600 may include: in step 610, the center point of the cross section of the tubular object in each cross-sectional image may be identified, and a plurality of radial lines directed toward the edge of the tubular object may be set with the center point as the starting point. In some embodiments, each point on the center line may be marked so that the position of the center point can be identified according to the mark when each cross-sectional image is detected. In some embodiments, the plurality of radial lines may be uniformly arranged, for example, the interval angle between each adjacent two radial lines in the plurality of radial lines may be the same. In other embodiments, the plurality of radial lines may be non-uniformly arranged, for example, the interval angle between each adjacent two radial lines in the plurality of radial lines may be different.

[0060] In some embodiments, the radiation line may be a straight line starting from the center point and emitting toward the edge of the tubular object. In other embodiments, the more radiation lines there are, the more accurate the detection result of the cross-sectional radius or cross-sectional diameter. In some other embodiments, the number of radiation lines may be an odd number or an even number. In some embodiments, the number of radiation lines may be set to 32 to 48. According to such a setting, the accuracy of the cross-sectional radius detection can be guaranteed, and the amount of data processing can be reduced to increase the data processing speed. Furthermore, compared with the case where less than 32 radiation lines are set, setting more than 32 radiation lines can be better applied to the measurement of the cross-sectional radius or cross-sectional diameter of irregular tubular cross sections.

[0061] Next, in step 620, the intersection of each radiation line and the edge of the tubular object in the cross-sectional image can be detected to determine the sampling point in the cross-sectional image. In some application scenarios, the cross-sectional image can include the cross section of the tubular object and the background image, and the radiation line can be extended into the background image so that the radiation line intersects with the edge of the tubular object to generate an intersection point. In other application scenarios, the radiation line can be extended only to the edge of the tubular object to intersect with the edge of the tubular object. In some embodiments, it can be determined that all intersection points are sampling points.

[0062] In other embodiments, step 620 may include: determining the intersection point on each radial line that is closest to the center point as the sampling point. According to such a setting, it is possible to effectively exclude the influence of the irregular tubular region that may exist in the cross-sectional image on the determination result of the cross-sectional radius or cross-sectional diameter of the tubular, for example, the irregular tubular region causes the situation that there are multiple sampling points on the same radial line, which is conducive to improving the accuracy of the detection result.

[0063] Then, the process can proceed to step 630, where the cross-sectional radius at the center point can be determined based on the average of the sampling radius between each sampling point and the center point. In some embodiments, the sampling radius can be the distance between the sampling point and the center point. In one embodiment, the cross-sectional radius can be calculated based on the following formula 2:

[0064]

[0065] Among them, r ij represents the cross-section radius, n represents the number of sampling points, represents the distance between the kth sampling point and the center point (i.e., the sampling radius), where f ij Represents the j-th cross-sectional image of the straightened tubular object i.

[0066] In another embodiment, the cross-sectional diameter can be calculated based on the following formula 3:

[0067]

[0068] Among them, d ij represents the cross-section diameter, n represents the number of sampling points, represents the distance between the kth sampling point and the center point (i.e., the sampling radius), where f ij Represents the j-th cross-sectional image of the straightened tubular object i.

[0069] It can be understood that the above implementation of determining the cross-sectional radius is exemplary and not restrictive. In one embodiment, detecting the cross-sectional radius or cross-sectional diameter at each center point may include: identifying the center point of the tubular cross section in each cross-sectional image, and setting a plurality of cross lines on the cross-sectional image with the center point as the intersection point; detecting the intersection of each cross line with the edge of the tubular in the cross-sectional image to determine the sampling point in the cross-sectional image; and determining the cross-sectional radius or cross-sectional diameter at the center point based on the average value of the sampling radius between each sampling point and the intersection point.

[0070] In some embodiments, the spacing angles between multiple cross lines may be the same or different. In other embodiments, the cross lines may be straight lines passing through the intersection. The more cross lines there are, the more accurate the detection result of the cross-sectional radius or cross-sectional diameter. In some further embodiments, the number of cross lines may be set to 8 to 16. The technical effect of setting 8 to 16 cross lines is similar to the technical effect of setting the number of 32 to 48 cross lines mentioned above, and will not be repeated here. In another embodiment, determining the sampling points in the cross-sectional image may include: dividing each cross line into two cross sub-lines at the intersection; and determining the intersection point on each cross sub-line that is closest to the intersection as the sampling point.

[0071] Compared with the scheme of setting cross lines, the operation of setting radial lines is simpler and more flexible. For example, an odd number of sampling points can be obtained by setting an odd number of radial lines, and the sampling points may not be symmetrical, so that in some application scenarios, the sampling points can be more representative. Compared with the scheme of setting radial lines, the scheme of setting cross lines can directly determine the cross-sectional diameter at the center point based on the average value of the distance between two sampling points on each cross line, without first determining the cross-sectional radius and then determining the cross-sectional diameter, and the cross-sectional diameter directly determined based on the cross lines will be more accurate.

[0072] like Figure 6As further shown in FIG. 6 , in another embodiment, method 600 may further include: in step 640 (shown in a dotted box), the relative difference between each sampling radius and the cross-sectional radius of the cross-sectional image may be calculated. In some embodiments, the relative difference of each sampling radius may be determined based on the difference between each sampling radius and the cross-sectional radius obtained in step 630. In other embodiments, the relative difference may be obtained by the ratio between the absolute value of the difference between the sampling radius and the cross-sectional radius and the cross-sectional radius. For example, the relative difference between each sampling radius and the cross-sectional radius of the cross-sectional image may be calculated by the following formula:

[0073]

[0074] in, It is used to represent the relative difference between the sampling radius of the kth sampling point in the jth cross-sectional image of the straightened tubular object i and the cross-sectional radius. represents the sampling radius of the kth sampling point, r ij Represents the cross-sectional radius of the j-th cross-sectional image of the straightened tubular object i.

[0075] Next, in step 650 (shown by the dashed box), in response to the relative difference being greater than a preset threshold, the sampling radius having the relative difference greater than the preset threshold may be determined as an abnormal sampling radius. In some embodiments, the preset threshold may be determined according to a multiple of the cross-sectional radius.

[0076] Further, in step 660 (shown in a dotted box), the cross-sectional radius may be updated according to the average value of the other sampling radii except the abnormal sampling radius. In one embodiment, the updated cross-sectional radius may be calculated based on the following formula:

[0077]

[0078] Among them, a ij It is used to represent the updated cross-sectional radius of the j-th cross-sectional image of the straightened tubular object i, m represents the updated number of samples, It represents the sampling radius of the kth sampling point in the jth cross-sectional image of the straightened tubular object i, excluding the abnormal sampling radius.

[0079] Combination of the above Figure 6The method for detecting the cross-sectional radius according to an embodiment of the present invention has been described in detail. It can be understood that, according to such an operation of updating the cross-sectional radius, the influence of the local abnormal features of the tubular object (such as adhesion, abnormal protrusions, burrs, etc.) on the detection result of the cross-sectional radius of the tubular object can be effectively eliminated, which is conducive to further improving the accuracy of the detection result of the cross-sectional radius. It can also be understood that the above description is exemplary and not restrictive. For example, in another embodiment of the present invention, the method according to the embodiment of the present invention may also include: based on the abnormal sampling radius and the updated cross-sectional radius, determining whether there is an abnormal protrusion at the abnormal sampling radius. The following will be combined with Figure 7 An exemplary description is given.

[0080] Figure 7 FIG. 2 is a flow chart showing a method for determining abnormal protrusions according to an embodiment of the present invention. Figure 7 As shown in , method 700 may include: in step 710, a first length at the abnormal sampling radius may be determined according to the difference between the maximum sampling radius of one or more adjacent abnormal sampling radii and the updated section radius. In some application scenarios, when the same cross-sectional image contains multiple abnormal protrusions, abnormal protrusions are judged according to one or more adjacent abnormal sampling radii, which can avoid the phenomenon of missing abnormal protrusions, so that the position and parameters of each abnormal protrusion can be accurately located.

[0081] In some embodiments, one or more adjacent abnormal sampling radii can be determined by judging whether abnormal sampling radii appear continuously in the same cross-sectional image. In other embodiments, adjacent abnormal sampling radii refer to that both sides of the abnormal sampling radius are normal sampling radii, or there is no normal sampling radius between multiple abnormal sampling radii. In other embodiments, the maximum sampling radius can be the maximum value of one or more abnormal sampling radii. In other embodiments, the first length can be calculated based on the following formula:

[0082]

[0083] Here L ij represents the first length in the j-th cross-sectional image of the straightened tubular object i, a ij It is used to indicate the updated section radius, c indicates the number of adjacent abnormal sampling radii, represents the j-th cross-sectional image f of the straightened tubular object i ij The b-th anomaly sampling radius in , Indicates the maximum sampling radius.

[0084] Next, in step 720, the first width at the abnormal sampling radius can be determined according to the width of the area where the maximum sampling radius exceeds the updated cross-sectional radius. In some embodiments, a boundary threshold of the area can be determined, and the portion where the maximum sampling radius exceeds the boundary threshold is determined as the above-mentioned area; based on the maximum width in the area, the first width of the area (i.e., the first width at the abnormal sampling radius) is determined. In other embodiments, a certain proportion of the updated cross-sectional radius can be determined as the boundary threshold as needed. For ease of understanding, a specific example is described below.

[0085] First, the radial line (or maximum sampling ray) where the maximum sampling radius is located can be determined. The method for determining the maximum sampling ray can be obtained based on the following formula:

[0086]

[0087] Among them, q ij It is used to indicate the radial line where the maximum sampling radius is located, c indicates the number of adjacent abnormal sampling radii, represents the j-th cross-sectional image f of the straightened tubular object i ij The b-th anomaly sampling radius in .

[0088] Then, the sampling ray q can be detected ij In the case of greater than λa ij The maximum width in the region is taken as the first width of the region, where λ represents the proportional coefficient, λa ij is the boundary threshold. In some embodiments, λ may be greater than 1, so as to avoid the position of the first width falling into the non-abnormal protrusion area. In other embodiments, λ may be 1.05-1.25.

[0089] Further, in step 730, in response to the first length being greater than the first length threshold and the first width being greater than the first width threshold, it can be determined that there is an abnormal protrusion at the abnormal sampling radius. The first length threshold and the first width threshold can be set as needed. In order to facilitate understanding of the determination method of the abnormal protrusion, the following will be combined with Figure 8 Further description is given.

[0090] Figure 8 2 is a schematic cross-sectional view showing the presence of abnormal protrusions according to an embodiment of the present invention. Figure 8 As shown in FIG. 8 , the cross-sectional image 800 includes a tubular cross section 801 of a straightened tubular object. By setting a plurality of radial lines 802 (shown by dashed lines) at the center point, a plurality of sampling points and a plurality of sampling radii can be determined, wherein for example Figure 6The method 600 shown in FIG. 1 determines multiple abnormal sampling radii (e.g., 803, 804, and 805 in the figure). Then, the first length can be determined according to the difference between the maximum sampling radius 805 among the abnormal sampling radii 803, 804, and 805 and the updated cross-sectional radius. Next, the first length can be determined according to λ times the updated cross-sectional radius (i.e., λa ij ) is located at position 806 (i.e., boundary threshold), and the maximum sampling radius 805 is determined to exceed λa ij The partial radius 808 of the position 806 can be determined, so that the area 807 where the partial radius 808 is located can be determined.

[0091] like Figure 8 As further shown in, the method for determining the first width 809 may include: setting multiple vertical lines on the partial radius 808, and detecting the intersection points of the multiple vertical lines with the edge of the area 807; determining the vertical line lengths of the multiple vertical lines in the area 807 based on the distance between the two intersection points on each vertical line; and determining the first width 809 based on the maximum value of the multiple vertical line widths.

[0092] Combination of the above Figure 7 and Figure 8 The method for determining abnormal protrusions according to an embodiment of the present invention has been described exemplarily. It can be understood that the determination of the abnormal sampling radius can not only be used to update the cross-sectional radius to obtain a more accurate cross-sectional radius, but also can be used to determine whether there is an abnormal protrusion in the cross-sectional image. According to such a setting, the method according to the embodiment of the present invention can not only accurately detect the stenosis rate of the tubular object, but also detect whether there is an abnormal protrusion in the tubular object, so that more morphological features of the tubular object can be more comprehensively evaluated, and more valuable data support can be provided for determining the state of the tubular object. In order to make it easier to understand the technical effect of the method for determining the cross-sectional radius according to an embodiment of the present invention, the following will be combined with Fig. 9 and Fig.10 Provide explanation.

[0093] Fig. 9 Schematic diagram showing the arrangement of cross lines according to an embodiment of the present invention. Taking the tubular object as a blood vessel as an example, Fig. 9 Figures (a), (b), (c), (d), and (e) show cross-sectional images of blood vessels of different shapes. Fig. 9As can be seen from Figures (a), (b), (c), (d), and (e), multiple intersection lines can be set to pass through the center point, and sampling points (such as the light-colored dots in the figure) can be determined based on the intersections of the intersection lines and the edges of the blood vessels. The cross-sectional radius of the blood vessel can be determined based on the distance between the sampling points and the center point, and the cross-sectional diameter of the blood vessel can also be determined based on the distance between two sampling points on the same intersection line. By determining the intersection point on each intersection sub-line that is closest to the intersection point as the sampling point, it is possible to effectively exclude, for example, Fig. 9 (d) shows the influence of the intersection of the cross line and the non-vascular area or the vascular adhesion area (such as the dark dots in the figure) on the determination result of the cross-sectional radius of the blood vessel.

[0094] It is understood that the visual effect of setting the radiation line in the cross-sectional image is different from Fig. 9 The visual effect of setting cross lines shown in FIG. 1 is similar to that of setting cross lines, and will not be described here. Compared with other methods for determining the cross-sectional radius, the technical solution of setting cross lines or setting radial lines according to the embodiment of the present invention is not only conducive to improving the accuracy of the measurement result, but also can be more widely applied to the measurement of the cross-sectional radius or cross-sectional diameter of the irregularly shaped tubular cross section caused by the tubular itself or image processing errors. Fig.10 Provide explanation.

[0095] Fig.10 is a comparison chart showing various cross-sectional radius measurement methods. Fig.10 As shown in , the first row shows, for example Fig. 9 The cross-sectional image shown in . The minimum distance method is a method of determining the distance between the point on the edge of the tubular object closest to the center point and the center point as the cross-sectional radius. The method is a method for calculating the cross-sectional radius of a tubular object based on the cross-sectional area of ​​the tubular object. Iteration is In addition, the cross-line setting method and the radial line setting method in the embodiment of the present invention can be collectively referred to as the multi-pass measurement method. Fig.10 It can be clearly seen that the multipath measurement method of the embodiment of the present invention can accurately measure the cross-sectional radius of blood vessel cross sections of various shapes, while other methods cannot achieve the applicable scope of the multipath measurement method of the embodiment of the present invention. In some application scenarios, setting two cross lines may be more suitable for detecting the cross-sectional radius of a circular blood vessel cross section. In comparison, setting more than 8 cross lines or more than 32 radial lines will significantly improve the reliability and accuracy of the measurement results for non-circular blood vessel cross sections.

[0096] In another embodiment, determining the average radius or average diameter may include: performing a weighted average operation on each cross-sectional radius according to the distance between the center point where the minimum cross-sectional radius among each cross-sectional radius in the straightened image is located and each center point to determine the average radius; or performing a weighted average operation on each cross-sectional diameter according to the distance between the center point where the minimum cross-sectional diameter among each cross-sectional diameter in the straightened image is located and each center point to determine the average diameter. In one embodiment, the average radius may be calculated based on the following formula 8:

[0097]

[0098] in, represents the mean radius, ω ij represents the weight value of the cross-sectional radius of the tube in the cross-sectional image at the jth center point of the straightened tube i, r ij Represents the cross-sectional radius of the tubular object in the cross-sectional image at the jth center point of the straightened tubular object i. ω ij It can be calculated based on the following formula nine:

[0099] ω ij =φ(D ij ) (Formula 9)

[0100] Where φ represents D ij The function of D ij The distance between the center point where the minimum cross-sectional radius is located in the straightened image of the straightened tubular object i and the jth center point can be expressed as the following formula 10:

[0101]

[0102] Among them, x ij represents the horizontal coordinate of the jth center point of the straightened tubular object i, The horizontal coordinate of the center point of the smallest cross-sectional radius of the straightened tubular object i. Fig.11 As shown in i1 represents the horizontal coordinate of the first center point of the straightened tubular object i, represents the horizontal coordinate of the lth center point of the straightened tubular object i, D ij represents the distance between the section where the minimum cross-sectional radius is located and the section where the j-th center point is located. In other embodiments, the calculation method of the average diameter may be similar to the calculation method of the average radius, which will not be repeated here.

[0103] In one embodiment, performing weighted average operation may include: constructing a distance function (e.g., φ) according to a preset rule; determining a weight value of each cross-sectional radius or each cross-sectional diameter based on the function; and performing weighted average operation on each cross-sectional radius or each cross-sectional diameter according to the weight value. In other embodiments, the preset rule may include: distance D ij The larger the center point, the smaller the corresponding weight value, for example Fig.12 As shown in oh i1 represents the weight value of the cross-sectional radius of the cross-sectional image of the straightened tubular object i at the first center point, Represents the weight value of the minimum cross-section radius, oh il The weight value of the cross-sectional radius of the cross-sectional image of the straightened tubular object i at the l-th center point. Through such a preset rule, the influence of the distance from the center point of the minimum cross-sectional radius or the minimum cross-sectional diameter on the weight value can be considered when determining the average radius or the average diameter, which is not only conducive to improving the accuracy of the calculation results, but also can more objectively reflect the real state of the tubular object.

[0104] In some other embodiments, the function may include at least one of a linear function, a quadratic function, an inverse proportional function, and a Gaussian function. ij The weight value of the cross-sectional radius at the j-th center point of the straightened tubular object i, a linear function of the distance can be obtained, for example, by the following formula 11:

[0105]

[0106] in, Indicates the maximum value among the distances of the cross-sectional radii of the cross-sectional images at the 1st to 1st center points of the straightened tubular object i.

[0107] In another embodiment, the quadratic function of the distance may be obtained by, for example, the following formula twelve:

[0108]

[0109] In yet another embodiment, the inverse proportional function of the distance may be obtained by, for example, the following formula 13:

[0110]

[0111] In one embodiment, the Gaussian function of the distance can be obtained by, for example, the following formula 14:

[0112]

[0113] The value of k can be 3.

[0114] Combination of the above Fig.11 and Fig.12 An exemplary description is given of a method for determining the average radius or average diameter of a tubular object according to an embodiment of the present invention. Those skilled in the art will appreciate that the above description is exemplary rather than restrictive. For example, the above-mentioned linear function, quadratic function, inverse proportional function and Gaussian function are exemplary. In some embodiments, other function forms that meet preset rules may be set as needed.

[0115] In one experimental example, by performing stenosis rate calculation based on the computer implementation of the embodiment of the present invention on the straightened tubular objects in 1665 straightened images, and comparing the stenosis rate results with the detection results of the imaging physician, it was found that the stenosis rate results obtained based on the embodiment of the present invention were highly consistent with the detection results of the imaging physician, indicating that the method according to the embodiment of the present invention has a high detection accuracy. In another experimental example, by using a computer to implement the method according to the embodiment of the present invention, the detection time for each straightened image is only about 2 seconds, and the stenosis rate of the tubular object in the straightened image can be determined, indicating that the method according to the embodiment of the present invention has a high detection speed.

[0116] In another experimental example, by determining the average radius or average diameter of the straightened images of 7 tubular objects according to the start-end point method, the average method and the weighted average method, and then calculating the average error of the stenosis rate results, the following can be obtained: Fig.13 The starting and ending point method may be a method for determining the average radius or average diameter based on the cross-sectional radius or cross-sectional diameter at the starting and ending points of the straightened tubular object. The average method may be a method for directly determining the average radius or average diameter based on the average cross-sectional radius or cross-sectional diameter at each center point of the straightened tubular object. The weighted average method may be a method for determining the average radius or average diameter based on the average cross-sectional radius or cross-sectional diameter at each center point of the straightened tubular object. Figure 1 The method shown in .

[0117] like Fig.13 As shown in , compared with the average error of 0.0837 for determining the stenosis rate by the start-end point method and the average error of 0.0534 for determining the stenosis rate by the average method, the average error of the stenosis rate determined by the weighted average method according to the embodiment of the present invention is only 0.0210. Therefore, the weighted average method according to the present invention can significantly reduce the average error of the stenosis rate detection of the tubular object. Through the above description of the technical solution for determining the stenosis rate of the tubular object implemented by the computer of the present invention and multiple embodiments, it can be understood by those skilled in the art that the method of the embodiment of the present invention can obtain the cross-sectional images at each center point in the straightened image, detect the cross-sectional radius or cross-sectional diameter of the tubular object based on each cross-sectional image, and determine the stenosis rate of the tubular object based on the weighted average radius or average diameter, and has the characteristics of fast detection speed, low average error and high accuracy.

[0118] In some embodiments, the cross-sectional radius can be detected by setting multiple radial lines on the cross-sectional image, which also has good applicability and detection accuracy for the detection of irregularly shaped tubular cross-sections. In other embodiments, the weighted average calculation of each cross-sectional radius can be performed based on the distance between the center point of the minimum cross-sectional radius of each cross-sectional radius in the straightened image and each center point, so as to obtain a more accurate average radius of the tubular object that is more in line with objective laws.

[0119] In a second aspect of the present invention, a device for determining the stenosis rate of a tubular object is provided, which may include: a cross-sectional image acquisition module, for acquiring, in response to receiving a straightened image of the tubular object, a cross-sectional image at each center point of the straightened tubular object in the straightened image; a detection module, for detecting the cross-sectional radius or cross-sectional diameter of the tubular cross section in the cross-sectional image at each center point; a mean determination module, for determining the average radius or average diameter of the straightened tubular object based on a weighted average of the cross-sectional radii or cross-sectional diameters at each center point; and a stenosis rate determination module, for determining the stenosis rate of the tubular object based on the minimum cross-sectional radius and the average radius among the cross-sectional radii, or based on the minimum cross-sectional diameter and the average diameter among the cross-sectional diameters.

[0120] The device of the embodiment of the present invention has been combined with Figure 1-Figure 12 Any of the described methods have been described and explained in detail and will not be repeated here.

[0121] In a third aspect of the present invention, there is provided an apparatus for determining the stenosis rate of a tubular object, comprising at least one processor; a memory storing program instructions, which, when executed by the at least one processor, causes the apparatus to execute any one of the methods described in the first aspect of the present invention. Fig.14 An exemplary description is given.

[0122] Fig.14 1400 is a schematic diagram of a system for determining the stenosis rate of a tubular object according to an embodiment of the present invention. The system 1400 may include a device 1410 according to an embodiment of the present invention, as well as its peripheral devices and an external network, wherein the device 1410 performs a stenosis rate detection operation for a straightened tubular object in a straightened image, so as to achieve the aforementioned combination Figure 1-Figure 13 The technical solution of any of the embodiments of the present invention.

[0123] like Fig.14As shown in , the device 1410 may include a CPU 1411, which may be a general-purpose CPU, a dedicated CPU, or other information processing and program execution unit. Further, the device 1410 may also include a large-capacity memory 1412 and a read-only memory ROM 1413, wherein the large-capacity memory 1412 may be configured to store various types of data, including centerline data, transformation matrix, etc., and various programs required for determining the stenosis rate of tubular objects, and the ROM 1413 may be configured to store the initialization of various functional modules in the system of the device 1410, the basic input / output driver of the system, and the data required for booting the operating system.

[0124] Furthermore, the device 1410 may also include other hardware or components, such as a graphics processor (“GPU”) 1415 and a field programmable gate array (“FPGA”) 1416, etc. It is understood that although a variety of hardware or components are shown in the device 1410, they are merely exemplary and not restrictive, and those skilled in the art may add or remove corresponding hardware according to actual needs.

[0125] The device 1410 of the embodiment of the present invention may further include a communication interface 1418, so that it can be connected to a local area network / wireless local area network (LAN / WLAN) 1450 through the communication interface 1418, and then connected to a local server 1460 or to the Internet ("Internet") 1470 through the LAN / WLAN. Alternatively or additionally, the device 1410 of the embodiment of the present invention may also be directly connected to the Internet or a cellular network through the communication interface 1418 based on wireless communication technology, such as wireless communication technology based on the third generation ("3G"), the fourth generation ("4G") or the fifth generation ("5G") . In some application scenarios, the device 1410 of the embodiment of the present invention may also access a server 1480 of an external network and a possible database 1490 as needed to obtain various known data and modules, etc., and may remotely store various detected data.

[0126] The peripheral devices of the device 1410 may include a display device 1420, an input device 1430, and a data transmission interface 1440. In one embodiment, the display device 1420 may, for example, include one or more speakers and / or one or more visual displays, which are configured to perform voice prompts and / or image video display on the detection process or final result of the device of the embodiment of the present invention. The input device 1430 may include, for example, a keyboard, a mouse, a microphone, a gesture capture camera, or other input buttons or controls, which are configured to receive input or user instructions of the detection information. The data transmission interface 1440 may include, for example, a serial interface, a parallel interface or a universal serial bus interface ("USB"), a small computer system interface ("SCSI"), a serial ATA, a firewire ("FireWire"), a PCI Express, and a high-definition multimedia interface ("HDMI"), etc., which are configured to transmit and interact with data of other devices or systems. According to the solution of the present invention, the data transmission interface 1440 can receive a straightened image, etc., and transmit various types of data and results to the device 1410.

[0127] The CPU 1411, mass storage 1412, read-only memory ("ROM") 1413, GPU 1415, FPGA 1416, and communication interface 1418 of the device 1410 of the embodiment of the present invention can be interconnected through a bus 1419, and data interaction with peripheral devices can be achieved through the bus. In one embodiment, through the bus 1419, the CPU 1411 can control other hardware components in the device 1410 and its peripheral devices.

[0128] In operation, the processor CPU 1411 of the device 1410 of the embodiment of the present invention can receive data through the input device 1430 or the data transmission interface 1440, and call the computer program instructions or codes (such as codes for determining the stenosis rate of tubular objects) stored in the memory 1412 to detect the received straightened image and its detection request, so as to obtain the stenosis rate of the straightened tubular object in the straightened image. Then, the processor CPU 1411 starts to perform operations such as acquiring a cross-sectional image, detecting a cross-sectional radius or a cross-sectional diameter, determining an average radius or an average diameter, and determining the stenosis rate of the tubular object according to the obtained straightened image. After the CPU 1411 obtains the stenosis rate of the tubular object by executing the program for determining the stenosis rate of the tubular object, the stenosis rate determination result can be displayed on the display device 1420 or output by voice prompt. In addition, the device 1410 can also upload the detection result to a network, such as a remote database 1490, through the communication interface 1418.

[0129] It should also be understood that any module, unit, component, server, computer, terminal or device that executes instructions of the present invention examples may include or otherwise access computer-readable media, such as storage media, computer storage media or data storage devices (removable) and / or non-removable) such as disks, optical disks or tapes. Computer storage media may include volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules or other data.

[0130] In a fourth aspect of the present invention, a computer-readable storage medium is provided, which stores a computer-implemented program for determining the stenosis rate of a tubular object. When the program is executed by a processor, the method according to any one of the first aspects of the present invention is executed.

[0131] The computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions that may be stored or otherwise maintained by such a computer-readable medium.

[0132] Further, the method for determining the stenosis rate of a tubular object implemented by a computer according to an embodiment of the present invention can also be applied to the assessment of the risk of cognitive dysfunction. For example, in the fifth aspect of the present invention, a device for predicting the risk of cognitive dysfunction is provided, which may include: a stenosis rate determination module, used to determine the stenosis rate of each cerebral artery according to any method described in the first aspect of the present invention; a binarization processing module, used to perform binarization processing on the stenosis rates of multiple cerebral arteries; a dimensionality reduction module, used to perform feature dimensionality reduction processing on the processing results of the binarization processing; and a classification model, used to classify the dimensionality reduction results after the feature dimensionality reduction processing, so as to predict the risk of cognitive dysfunction in the owner of multiple cerebral arteries.

[0133] In some embodiments, the binarization process may include: in response to a stenosis rate of a certain cerebral artery being greater than or equal to a first threshold, the cerebral artery may be determined to be a stenotic vessel; in response to a stenosis rate of a certain cerebral artery being less than the first threshold, the cerebral artery may be determined to be a non-stenotic vessel. In other embodiments, the binarization process may further include at least one of the following: marking a cerebral artery determined to be a stenotic vessel as 1; marking a cerebral artery determined to be a non-stenotic vessel as 0.

[0134] In other embodiments, the processing results of the binarization processing can be refined by feature dimensionality reduction, for example, using t-SNE technology, so as to highlight the difference features that are helpful for predicting cognitive dysfunction. t-SNE (t-distributed stochastic neighbor embedding) is a machine learning algorithm for dimensionality reduction processing. In some other embodiments, the classification model can be implemented by at least one of the machine learning models support vector machine (such as Kernel-SVM or Linear-SVM), multi-layer perceptron (MLP), DBSCAN clustering algorithm, etc., to classify the dimensionality reduction results to predict the risk of cognitive dysfunction. In some embodiments, the classification model can directly classify the binarization results to predict the risk of cognitive dysfunction. In an experimental example, the device for predicting the risk of cognitive dysfunction according to an embodiment of the present invention was used to train and test 185 subjects using 5-fold cross validation, and a prediction accuracy of more than 90% can be achieved.

[0135] Furthermore, the method for determining abnormal protrusions according to an embodiment of the present invention can also be applied to assessing the risk level of aneurysm rupture. For example, in a sixth aspect of the present invention, a device for assessing the risk level of aneurysm rupture can be provided, which can include: an abnormal protrusion determination module, which is used to determine the risk level of aneurysm rupture according to, for example, Figure 7 The method shown in determines whether there is an abnormal protrusion in each cross-sectional image of a tubular object, wherein the tubular object can be an arterial blood vessel; an aneurysm determination module, which can be used to determine the maximum value of the first length in the multiple cross-sectional images as the second length at the abnormal protrusion in response to the presence of the abnormal protrusion in multiple consecutive cross-sectional images, and determine the maximum value of the first width in the multiple cross-sectional images as the second width at the abnormal protrusion; and in response to the second length being greater than a second length threshold and the second width being greater than a second width threshold, determine that the abnormal protrusion is an aneurysm on a cerebral arterial blood vessel; an evaluation module, which can be used to evaluate the rupture risk level of the aneurysm based on the second length and the second width.

[0136] In some embodiments, evaluating the rupture risk level of an aneurysm may include: in response to the second length being less than a third length threshold, and the second width being less than a third width threshold, the aneurysm may be determined to have a low rupture risk; in response to the second length being greater than the third length threshold and less than a fourth length threshold, and in response to the second width being greater than the third width threshold and less than a fourth width threshold, the aneurysm may be determined to have a medium rupture risk; in response to the second length being greater than the fourth length threshold, and the second width being greater than the fourth width threshold, the aneurysm may be determined to have a high rupture risk.

[0137] Although the embodiments of the present invention are described above, the contents are only embodiments used to facilitate understanding of the present invention, and are not intended to limit the scope and application scenarios of the present invention. Any technician in the technical field of the present invention can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed by the present invention, but the scope of patent protection of the present invention shall still be based on the scope defined by the attached claims.

Claims

1. A computer-implemented method for determining the stenosis rate of a tubular object, characterized in that: include: In response to receiving the straightened image of the tubular object, acquiring cross-sectional images at each center point of the straightened tubular object in the straightened image; Detecting the cross-sectional radius or cross-sectional diameter of the cross-sectional image of each center point; performing a weighted average operation on each cross-sectional radius according to the distance between the center point where the minimum cross-sectional radius among each cross-sectional radius in the straightened image is located and each center point, so as to determine the average radius; or performing a weighted average operation on each cross-sectional diameter according to the distance between the center point where the minimum cross-sectional diameter among each cross-sectional diameter in the straightened image is located and each center point, so as to determine the average diameter; and The stenosis rate of the tubular object is determined based on the minimum cross-sectional radius among the cross-sectional radii and the average radius, or based on the minimum cross-sectional diameter among the cross-sectional diameters and the average diameter.

2. The method according to claim 1, characterized in that Detecting the cross-section radius at each center point includes: Identify the center point of the cross section of the tubular object in each cross-sectional image, and set a plurality of radial lines toward the edge of the tubular object with the center point as the starting point; Detecting the intersection of each radiation line with the edge of the tubular object in the cross-sectional image to determine a sampling point in the cross-sectional image; and The cross-sectional radius at the center point is determined according to the average value of the sampling radius between each sampling point and the center point.

3. The method according to claim 2, characterized in that Also includes: Calculate the relative difference between each sampling radius and the section radius of the section image where it is located; In response to the relative difference being greater than a preset threshold, determining the sampling radius having the relative difference greater than the preset threshold as an abnormal sampling radius; as well as The cross-sectional radius is updated according to an average value of other sampling radii except the abnormal sampling radius.

4. The method according to claim 3, characterized in that: Also includes: Determine a first length at the abnormal sampling radius according to a difference between a maximum sampling radius among one or more adjacent abnormal sampling radii and an updated cross-sectional radius; Determining a first width at the abnormal sampling radius according to a width of an area where the maximum sampling radius exceeds an updated cross-sectional radius; as well as In response to the first length being greater than a first length threshold and the first width being greater than a first width threshold, it is determined that an abnormal protrusion exists at the abnormal sampling radius.

5. The method according to claim 1, characterized in that The weighted average operation includes: Constructing a function of the distance according to a preset rule; Determine a weight value of each cross-sectional radius or each cross-sectional diameter based on the function; and A weighted average calculation is performed on each cross-sectional radius or each cross-sectional diameter according to the weight value.

6. The method according to any one of claims 1 to 5, characterized in that: The tubular object includes a cerebral arterial blood vessel.

7. A device for determining the stenosis rate of a tubular object, characterized in that: include: A cross-sectional image acquisition module, configured to acquire, in response to receiving a straightened image of the tubular object, a cross-sectional image at each center point of the straightened tubular object in the straightened image; A detection module, used for detecting the cross-sectional radius or cross-sectional diameter of the cross-sectional image of the tubular object at each center point; a mean value determination module, configured to perform a weighted average operation on each cross-sectional radius based on a weighted average value of the cross-sectional radius or cross-sectional diameter at each center point, according to the distance between the center point where the minimum cross-sectional radius among each cross-sectional radius in the straightened image is located and each center point, so as to determine the mean radius; or perform a weighted average operation on each cross-sectional diameter based on the distance between the center point where the minimum cross-sectional diameter among each cross-sectional diameter in the straightened image is located and each center point, so as to determine the mean diameter; as well as The stenosis rate determination module is used to determine the stenosis rate of the tubular object based on the minimum cross-sectional radius among the cross-sectional radii and the average radius, or based on the minimum cross-sectional diameter among the cross-sectional diameters and the average diameter.

8. A device for determining the stenosis rate of a tubular object, characterized in that: include, at least one processor; A memory storing program instructions, which, when executed by the at least one processor, causes the device to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The device stores a computer-implemented program for determining the stenosis rate of a tubular object. When the program is run by a processor, the method according to any one of claims 1 to 6 is performed.

10. A device for predicting the risk of cognitive impairment, characterized in that: include: a stenosis rate determination module, used to determine the stenosis rate of each cerebral artery according to the method of claim 6; A binarization processing module, used for binarization processing of the stenosis rates of multiple cerebral arteries; A dimension reduction module, used for performing feature dimension reduction processing on the processing result of the binarization processing; as well as The classification model is used to classify the dimension reduction results after the feature dimension reduction processing to predict the risk of cognitive dysfunction in the owner of the multiple cerebral arteries.

Images