An auxiliary path planning method for inferior vena cava filter implantation puncture target point

Abstract

The present application relates to the technical field of image analysis, and in particular to a kind of inferior vena cava filter implantation puncture target point auxiliary path planning method.Multiple CT images of the inferior vena cava of the patient are obtained, and the region to be distinguished is preliminarily divided based on the shape change of the blood vessel.Then, combined with the shape features and area change of the region to be distinguished, the blood vessel overlap representation evaluation index is determined, and the overlap and non-overlap regions are distinguished.Next, the overlap and non-overlap regions are combined, the shape similarity and overlap representation evaluation index similarity are analyzed, the smooth continuation index is determined, and the matching combination is screened out based on the index and the proportion of blood vessel overlap area.In addition, the gray value difference of blood vessels at different levels in the CT image is also considered, and the non-blood vessel overlap region is evaluated at different levels to more accurately represent the region belonging to the same blood vessel.Finally, according to the hierarchical evaluation result, the navigation path planning of the puncture target point in the region with blood vessel overlap is more accurate, and the planning accuracy is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image analysis, in particular to an auxiliary path planning method for puncture target points of inferior vena cava filter placement. BACKGROUND

[0002] Inferior vena cava filter placement is an effective means to prevent pulmonary embolism caused by deep vein thrombosis in the lower extremities. When performing such surgery, it is crucial to accurately determine the puncture target point to ensure that the filter can be accurately placed in the predetermined inferior vena cava position. Due to the complexity of the inferior vena cava and its surrounding structures, and the anatomical features of different patients, to improve the accuracy and safety of placement, path planning assistance technology has emerged. Through modern imaging technology (such as CT, ultrasound, etc.) combined with computer-aided navigation system, the navigation path of the puncture target point during the operation can be planned in real time to avoid damage to the surrounding important blood vessels and organs.

[0003] When planning the navigation path of the puncture target point of the retrievable inferior vena cava filter, the existing technology usually relies on the shortest distance between the puncture target points to plan the navigation path. However, in the case of complex vascular structures and vascular overlap, simple shortest distance calculation cannot effectively avoid complex structures in such cases, reducing the accuracy of the auxiliary confirmation of the puncture target point of the retrievable inferior vena cava filter, and easily leading to misjudgment, affecting the accuracy of the final navigation path. SUMMARY

[0004] In order to solve the technical problem that in the case of complex vascular structures and vascular overlap, simple shortest distance calculation cannot effectively avoid complex structures in such cases, easily leading to misjudgment, and affecting the accuracy of the navigation path, the purpose of the present application is to provide an auxiliary path planning method for puncture target points of inferior vena cava filter placement, and the technical solution adopted is as follows:

[0005] Obtain multiple CT images of the inferior vena cava of a patient, and segment the blood vessel region in each CT image;

[0006] In each CT image, based on the distance change between the two edge lines of each blood vessel in the blood vessel region, the region to be divided is divided; the shape feature of each region to be analyzed in each CT image and the area change between each region to be analyzed and the corresponding region in other CT images are analyzed, and the blood vessel overlap representation evaluation index of each region to be divided is determined to distinguish the blood vessel overlap region and the non-blood vessel overlap region;

[0007] In each CT image, each blood vessel overlapping area is combined with a non-blood vessel overlapping area to obtain a plurality of area combinations; in each area combination, the shape similarity between the blood vessel overlapping area and the non-blood vessel overlapping area and the similarity between the blood vessel overlapping area and the non-blood vessel overlapping area are analyzed, and a smooth continuation index of the area combination is determined; the differences between the smooth continuation indexes of all area combinations are analyzed, and the proportion of the blood vessel overlapping area in the area combination is combined to screen a matching combination from all area combinations;

[0008] In each CT image, the differences in gray value between the non-blood vessel overlapping areas in the matching combination corresponding to each blood vessel overlapping area are analyzed, the non-blood vessel overlapping areas are hierarchically evaluated, and a hierarchical evaluation result is obtained; and a navigation path of a puncture target point is planned according to the hierarchical evaluation result.

[0009] Further, the method for obtaining the region to be distinguished comprises:

[0010] In each CT image, one of the two edge lines of each blood vessel is selected as a target line;

[0011] On the target line, a feature point is obtained based on a SURF feature point detection algorithm, and a straight line perpendicular to the target line is drawn through each feature point, and the intersection of the straight line with the other edge line is taken as a reference point corresponding to each feature point;

[0012] The distance between each feature point and the corresponding reference point is taken as a distance factor;

[0013] On the target line, a feature point not existing in the feature point set is randomly selected as a test point, and the test point corresponds to a feature point set;

[0014] Taking the test point as a starting point, the feature points on both sides of the target line are traversed respectively, and for each traversed feature point, the differences in distance factors between the traversed feature point and the feature points in the feature point set are analyzed to obtain a merging index; when the merging index meets a preset condition, the traversed feature point is merged into the feature point set of the test point, and the traversal is continued; if not, the traversal is stopped;

[0015] The preset condition is set to be greater than or equal to a preset merging threshold;

[0016] On the target line, a feature point not existing in the feature point set is randomly selected as a test point to repeat the above traversal process until all feature points exist in the feature point set, and all feature point sets on the target line are obtained;

[0017] In each feature point set, the two feature points farthest from each other on the target line are taken as end points, and the region surrounded by the line segment between the end points and the corresponding reference points, the line segment of the two end points on the edge line, and the line segment of the two end points on the target line is taken as a to-be-distinguished region.

[0018] Further, the method for obtaining the merging index comprises:

[0019] Taking each traversed feature point as a to-be-analyzed point;

[0020] Taking the absolute value of the difference between the distance factor of the to-be-analyzed point and the distance factor of each feature point in the feature point set corresponding to the to-be-measured point as a difference factor;

[0021] Taking the value obtained by negatively correlating and normalizing the mean of all the difference factors corresponding to the to-be-analyzed point as the merging index.

[0022] Further, the method for obtaining the blood vessel overlap representation evaluation index comprises:

[0023] Matching the feature points in different CT images, thereby determining the corresponding region of each to-be-distinguished region in each CT image in other CT images, and taking the corresponding region of each to-be-distinguished region in each CT image in other CT images as a contrast region;

[0024] In each to-be-distinguished region and all corresponding contrast regions, taking the difference between the maximum region area and the minimum region area as a first overlap representation evaluation factor corresponding to each to-be-distinguished region;

[0025] In each to-be-distinguished region, taking the line connecting the midpoint of the line segment between the feature point and the corresponding reference point as a center line, taking the mean of the curvature values at all the midpoints on each center line as a bending degree value, and taking the variance of the curvature values at all the midpoints on each center line as a fluctuation degree value;

[0026] Taking the value obtained by normalizing the product of the bending degree value and the fluctuation degree value corresponding to each to-be-distinguished region as a second overlap representation evaluation factor corresponding to each to-be-distinguished region;

[0027] Taking the value obtained by normalizing the sum of the first overlap representation evaluation factor and the second overlap representation evaluation factor corresponding to each to-be-distinguished region as the blood vessel overlap representation evaluation index of each to-be-distinguished region.

[0028] Further, the method for distinguishing the blood vessel overlap region and the non-blood vessel overlap region comprises:

[0029] When the blood vessel overlap representation evaluation index of a to-be-distinguished region is greater than a preset overlap threshold, the to-be-distinguished region is considered as a blood vessel overlap region.

[0030] When the blood vessel overlap representation evaluation index of a certain region to be distinguished is less than or equal to the preset overlap threshold value, the region to be distinguished is considered as a non-blood vessel overlap region.

[0031] Further, the method for obtaining the region combination comprises:

[0032] In each CT image, for any blood vessel overlap region, a non-blood vessel overlap region connected with the blood vessel overlap region is taken as a candidate region;

[0033] In two directions of the blood vessel overlap region along the blood vessel edge line, each optional candidate region and the blood vessel overlap region form a region combination, and all non-repeated region combinations are obtained.

[0034] Further, the method for obtaining the smooth continuation index comprises:

[0035] In each region combination, an intersection point between a straight line where a center line of each non-blood vessel overlap region is located and a straight line where a center line of the blood vessel overlap region is located is taken as a convergence point;

[0036] A Euclidean distance between a midpoint at a maximum curvature value on a center line corresponding to each non-blood vessel overlap region and a corresponding convergence point is taken as a position deviation factor, and a value obtained by negatively correlating and mapping a sum value of the position deviation factors corresponding to two non-blood vessel overlap regions is taken as a smooth factor;

[0037] A value obtained by negatively correlating and normalizing an absolute value of a difference between blood vessel overlap representation evaluation indexes of the two non-blood vessel overlap regions is taken as a similarity factor;

[0038] A value obtained by normalizing a sum value of the smooth factor and the similarity factor corresponding to each region combination is taken as a smooth continuation index of each region combination.

[0039] Further, the method for obtaining the matching combination comprises:

[0040] In each CT image, for any blood vessel overlap region, a proportion of a smooth continuation index of each region combination corresponding to the blood vessel overlap region in smooth continuation indexes of all region combinations corresponding to the blood vessel overlap region is taken as a first matching contribution degree;

[0041] In each region combination corresponding to the blood vessel overlap region, a length average of two edge lines of the blood vessel in the blood vessel overlap region is calculated as a first length factor, a length average of two edge lines of the blood vessel in each non-blood vessel overlap region is calculated as a second length factor, and a proportion of the first length factor in a sum value of the first length factor and all second length factors is taken as a second matching contribution degree.

[0042] The sum value of the first matching contribution and the second matching contribution of each region combination corresponding to the blood vessel overlapping region is normalized as a matching coefficient;

[0043] Among all region combinations of the blood vessel overlapping region, the region combination with a matching coefficient greater than a preset matching threshold is regarded as a matching combination.

[0044] Further, the method for obtaining the hierarchical evaluation result comprises:

[0045] Among all matching combinations, in all non-blood vessel overlapping regions along each direction of the blood vessel overlapping region, the mean value of the gray scale values of all pixel points in each non-blood vessel overlapping region is calculated as the gray scale feature value of each non-blood vessel overlapping region, all non-blood vessel overlapping regions are arranged in ascending order according to the gray scale feature values, and a sorting sequence is obtained.

[0046] In the two sorting sequences corresponding to the blood vessel overlapping region, the two non-blood vessel overlapping regions with the same sequence number value belong to the same level of blood vessel.

[0047] Further, the method for obtaining the blood vessel region comprises:

[0048] Each CT image is taken as the input of the pre-trained neural network, so as to output the blood vessel region in each CT image.

[0049] The present application has the following beneficial effects:

[0050] In the navigation path planning of the puncture target point of the inferior vena cava filter, the influence of the blood vessel overlap is analyzed in the present application, because the blood vessel overlap may exist and affect the planning accuracy of the navigation path. Firstly, a plurality of CT images of the inferior vena cava of a patient are obtained, and the blood vessel regions are segmented. Since the shape of the blood vessel changes when the blood vessels overlap, that is, the distance between the edge lines of the blood vessels changes obviously, the blood vessel regions are preliminarily divided based on this feature, and the regions to be divided are obtained. Further, when the blood vessels overlap, the area of the overlapping part may change in different CT images, so the shape feature of each region to be analyzed and the area change of the region to be analyzed and the corresponding region in other CT images are comprehensively considered to determine the blood vessel overlap representation evaluation index of the region to be divided, and the blood vessel overlap region and the non-blood vessel overlap region are divided based on the index. In order to determine which non-blood vessel overlap regions on both sides belong to the same blood vessel as the blood vessel overlap region, the blood vessel overlap region and the non-blood vessel overlap region are combined to obtain a plurality of region combinations. Since the spatial continuity and shape should be consistent and smooth if they belong to the same blood vessel, the shape similarity between the blood vessel overlap region and the non-blood vessel overlap region and the similarity between the overlap representation evaluation indexes of the non-blood vessel overlap regions are analyzed in each region combination to determine the smooth continuation index of the region combination. Since the larger the proportion of the blood vessel overlap region is, the smaller the proportion of the non-blood vessel overlap part is, and the matching difficulty is reduced, the weight should be increased, so based on the difference between the smooth continuation indexes and the proportion of the blood vessel overlap region, the matching combination is screened out in all region combinations, and the regions in each matching combination represent that they are more likely to belong to the same blood vessel. Further, since the gray values of the blood vessels at different levels in the CT images are different, the difference of the gray values between the non-blood vessel overlap regions of each matching combination corresponding to each blood vessel overlap region is analyzed in the present application, so that the level of the non-blood vessel overlap region is evaluated, and the level evaluation result is obtained. The level evaluation result can more accurately represent which regions belong to the same blood vessel, and the navigation path of the puncture target point can be more accurately planned according to the level evaluation result for the region with blood vessel overlap. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort.

[0052] Figure 1This is a flowchart of an auxiliary path planning method for inferior vena cava filter placement puncture target point provided in an embodiment of the present invention;

[0053] Figure 2 This is a schematic diagram illustrating the division of a vascular region according to an embodiment of the present invention;

[0054] Figure 3 This is a comparative schematic diagram of normal blood vessels and overlapping blood vessels provided in one embodiment of the present invention;

[0055] Figure 4 This is a schematic diagram illustrating the representation of a distance factor according to an embodiment of the present invention;

[0056] Figure 5 This is a schematic diagram of the division of a region to be distinguished according to an embodiment of the present invention;

[0057] Figure 6 This is a partial schematic diagram of a blood vessel overlapping region provided in one embodiment of the present invention;

[0058] Figure 7 This invention provides a navigation path planning process according to one embodiment of the present invention. Detailed Implementation

[0059] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an auxiliary path planning method for inferior vena cava filter placement puncture target point proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0061] The following describes in detail, with reference to the accompanying drawings, a specific scheme of the auxiliary path planning method for inferior vena cava filter placement puncture target point provided by the present invention.

[0062] Please see Figure 1 The diagram illustrates a method flowchart for an auxiliary path planning method for inferior vena cava filter placement puncture target point according to an embodiment of the present invention. The method includes the following steps:

[0063] Step S1: Acquire multiple CT images of the patient's inferior vena cava and segment the vascular region in each CT image.

[0064] The implantation operation of the retrievable inferior vena cava filter (IVC filter) is usually used for preventing and treating pulmonary embolism caused by deep vein thrombosis, and one of the key links in the implantation process is to accurately determine the puncture target point, that is, to select a suitable puncture position and accurately place the filter in the inferior vena cava. In order to improve the accuracy and safety of implantation, path planning auxiliary technology emerges as the times require. Through modern imaging technology (such as CT, ultrasound, etc.) combined with a computer-aided navigation system, the puncture target point can be accurately positioned in real time during the operation, and the feasibility of the puncture path can be evaluated to avoid damage to the surrounding important blood vessels and organs.

[0065] However, due to the complex structure of blood vessels and the possible existence of blood vessel overlap, the existing technology relies on the shortest distance between the puncture target points to plan the navigation path, which cannot effectively avoid the complex blood vessel structure, and the existence of blood vessel overlap area will affect the accuracy of the navigation path. Therefore, in the embodiment of the present application, the influence of the existing overlap area of the blood vessels on the navigation path planning of the puncture target point is mainly analyzed.

[0066] Firstly, a plurality of CT images of the inferior vena cava of the patient need to be obtained. Specifically, a high-resolution CT scanning device can be used to scan the inferior vena cava position of the patient. When performing CT scanning, the injection of iodine-containing contrast agent can help to enhance the contrast between blood vessels and surrounding tissues. The CT scanning device can obtain a plurality of CT images of the inferior vena cava of the patient, and each CT image can display different structures in the inferior vena cava of the patient, including blood vessels, organs and bones, etc.

[0067] It should be noted that in the scanning process, the CT images are taken once every second in the embodiment of the present application, and a plurality of CT images can be obtained after the scanning is completed.

[0068] Since there can be various structures in the obtained CT images, and the blood vessel region is mainly analyzed in the embodiment of the present application, the blood vessel region needs to be segmented in each CT image.

[0069] Preferably, in an embodiment of the present application, the method for obtaining the blood vessel region comprises:

[0070] Each CT image is taken as the input of the pre-trained neural network, so as to output the blood vessel region in each CT image; wherein when the blood vessel exists bifurcation, the blood vessel is recorded as two different blood vessels, that is, the blood vessel region can contain multiple blood vessels. Please refer to Figure 2 which shows the division of the blood vessel region in an embodiment of the present application.

[0071] It should be noted that the neural network in this embodiment of the present application can adopt CNN, and the training process of the neural network is a known technology, and the specific process will not be described here.

[0072] The collection and acquisition of the CT image data in the embodiments of the present application are all authorized by relevant users, the process does not violate relevant laws and regulations, and does not violate public order and good customs.

[0073] Step S2: In each CT image, based on the change condition of the distance between the two edge lines of each blood vessel in the blood vessel region, the to-be-differentiated region is divided; the shape feature of each to-be-differentiated region in each CT image and the area change condition between each to-be-analyzed region and the corresponding region in other CT images are analyzed, the blood vessel overlap representation evaluation index of each to-be-differentiated region is determined, so as to differentiate the blood vessel overlap region and the non-blood vessel overlap region.

[0074] When the navigation path of the puncture target point of the retrievable inferior vena cava filter is planned, the inferior vena cava region often has a complex blood vessel structure and blood vessel overlap. The overlap of these blood vessel structures will cause multiple blood vessel images to be interwoven in the CT image. Since the accuracy of the navigation path is directly related to the smooth progress of the inferior vena cava filter placement process, the influence of the blood vessel overlap will not only increase the ambiguity of the navigation path, but also may reduce the confirmation accuracy of the puncture target point, so it is necessary to differentiate the blood vessel overlap region in the CT image of the inferior vena cava in the path planning process of the puncture target point of the retrievable inferior vena cava filter.

[0075] Blood vessel overlap will cause the shape of the blood vessel in the CT image to change. Under normal circumstances, the diameter of the blood vessel gradually decreases from the beginning to the end. This change is usually smooth and uniform, ensuring that the blood can flow at the expected pressure and speed. In the part where no overlap occurs, the diameter of the blood vessel remains normal. However, when the blood vessel overlaps, especially when two or more blood vessels overlap or cross at a certain position, the diameter of the blood vessel will change locally. In the overlapping area, due to the overlap of the blood vessels, the diameter of the blood vessel will significantly widen. The transition of the diameter of the blood vessel from the overlapping area to the non-overlapping area becomes non-smooth and may have a sudden change, resulting in the increase and decrease of the diameter of the blood vessel being no longer uniform. Please refer to Figure 3 which shows a comparison diagram of normal blood vessels and overlapping blood vessels in an embodiment of the present application. In the CT image, the change in the diameter of the blood vessel can be represented by the change in the distance between the two edge lines of each blood vessel in the blood vessel region. Therefore, based on the foregoing features, the blood vessel region is divided to obtain the to-be-differentiated region, which may be a blood vessel overlap region or a non-blood vessel overlap region.

[0076] For the blood vessel overlapping region, in addition to the shape feature of the blood vessel, the area of the blood vessel at the overlapping position will present different sizes in different CT images, the blood vessel not overlapping only presents the area difference of the single blood vessel, and the area of the single blood vessel in different CT images will not change significantly, therefore, on the basis of the region to be distinguished, the shape feature of the region to be distinguished and the area change between each region to be distinguished and the corresponding region in other CT images are comprehensively considered, the blood vessel overlapping evaluation index of each region to be distinguished is calculated, the index can more accurately reflect whether the blood vessel overlaps, and therefore, based on the index, whether the region to be distinguished belongs to the blood vessel overlapping region can be judged.

[0077] Firstly, in each CT image, the region to be distinguished is divided based on the distance change between the two edge lines of each blood vessel in the blood vessel region. Preferably, in an embodiment of the present application, the method for obtaining the region to be distinguished comprises the following steps:

[0078] In each CT image, the edge line of each blood vessel in the blood vessel region can be obtained based on the Canny operator, and one of the two edge lines of each blood vessel is selected as the target line.

[0079] On the target line, the feature points are obtained based on the SURF feature point detection algorithm. The feature points are used to simplify the subsequent calculation steps, so that the division of the region to be distinguished can be performed by analyzing a few key positions. A straight line perpendicular to the target line is drawn through each feature point, and the intersection of the straight line with the other edge line is taken as the reference point corresponding to each feature point. At this time, the distance between each feature point and the corresponding reference point is taken as the distance factor, which can represent the diameter of the blood vessel at each feature point, and is used for subsequent quantification of the diameter change of the blood vessel. The greater the distance factor, the greater the diameter of the blood vessel, and vice versa. Please refer to Figure 4 which shows the representation diagram of the distance factor in an embodiment of the present application.

[0080] It should be noted that the Canny operator and the SURF feature point detection algorithm are both known technologies, and the specific process is not described here.

[0081] Then the diameter change of the blood vessel can be analyzed to divide the region to be distinguished, so that the diameter of the blood vessel in each region to be distinguished has a relatively consistent change.

[0082] In this embodiment of the present application, the division of the region to be distinguished is mainly performed by traversing the feature points:

[0083] On the target line, a feature point not existing in the feature point set is randomly selected as the test point, and the test point corresponds to a feature point set.

[0084] Taking the to-be-tested point as a starting point, the feature points on both sides of the target line are traversed respectively, for each traversed feature point, the difference of the distance factor between the feature point and the feature points in the feature point set is analyzed, and a merging index is obtained, when the merging index meets the preset condition, the traversed feature point is merged into the feature point set of the to-be-tested point, and the traversal is continued, if not, the traversal is stopped;

[0085] The preset condition is set to be greater than or equal to a preset merging threshold.

[0086] On the target line, a feature point not existing in the feature point set is randomly selected as a to-be-tested point to repeat the above traversal process, until all feature points exist in the feature point set, and the traversal is stopped, and all feature point sets on the target line are obtained.

[0087] The merging index includes:

[0088] Taking each traversed feature point as a to-be-analyzed point;

[0089] The absolute value of the difference between the distance factor of the to-be-analyzed point and the distance factor of each feature point in the feature point set corresponding to the to-be-tested point is taken as a difference factor, at this time, each feature point in the feature point set has a difference factor with the to-be-analyzed point, when the difference factor is larger, it means that there is a larger difference between the diameter of the blood vessel at the to-be-analyzed point and the diameter of the blood vessel at the feature point in the feature point set, then the larger the mean value of all difference factors is, the to-be-analyzed point should not be merged into the feature point set, on the contrary, the smaller the difference factor is, the higher the consistency between the diameter of the blood vessel at the to-be-analyzed point and the diameter of the blood vessel at the feature point in the feature point set is, then the smaller the mean value of all difference factors is, the to-be-analyzed point should be merged into the feature point set, therefore, the mean value of all difference factors corresponding to the to-be-analyzed point can be negatively correlated and normalized, so as to realize the logical relationship correction, and obtain the merging index, at this time, the larger the merging index is, the more the to-be-analyzed point should be merged into the feature point set. The negative correlation and normalization processing can adopt the formula exp(-x), wherein exp() represents the exponential function with natural constant e as the base, and x represents the independent variable.

[0090] Based on the foregoing process, the feature points on the target line of each blood vessel in the blood vessel region are divided, so that the continuous and relatively consistent feature points exist in the same feature point set, and finally in each feature point set, the two feature points farthest away on the target line are taken as end points, the line segment between the end points and the corresponding reference points, the line segment of the reference points on the edge line corresponding to the two end points, and the line segment of the two end points on the target line, and the region surrounded thereby is taken as a to-be-divided region.

[0091] It should be noted that the preset merging threshold is set to 0.65, and the specific value can be adjusted according to the implementation scene, which is not limited here.

[0092] Here, the acquisition process of the to-be-distinguished region is exemplified: on the target line, initially, all feature points are not in the feature point set, so a feature point 10 is randomly selected as a to-be-tested point, and the feature point 10 corresponds to a feature point set a, at this time, there is only the feature point 10 in the feature point set a. Taking the to-be-tested point 10 as the starting point, the feature points on both sides of the target line are traversed along the target line, first traversing in one direction along the target line, the traversed feature point is feature point 11, the merging indicator between the feature point 11 and the feature point 10 in the feature point set a is calculated, if the merging indicator is greater than or equal to the preset merging threshold, the feature point 11 is merged into the feature point set a, at this time, there are feature points 10 and 11 in the feature point set a; then the feature point 12 is continuously traversed, at this time, the merging indicator between the feature point 12, the feature point 10 and the feature point 11 is calculated, if the merging indicator is still greater than or equal to the preset merging threshold, the feature point 12 is also merged into the feature point set a, at this time, there are feature points 10, 11 and 12 in the feature point set a, the above process is continuously repeated, if the feature point 15 is traversed and the preset condition is not met, the process is stopped, at this time, there are feature points 10, 11, 12, 13 and 14 in the feature point set; then the to-be-tested point 10 is taken as the starting point, and the other direction along the target line is traversed, the specific process is the same as above, when the traversal in the direction is completed, the first feature point set a can be obtained, if at this time, there are feature points 7, 8, 9, 10, 11, 12, 13 and 14 in the feature point set a, then a feature point is continuously selected as a to-be-tested point from other feature points except the feature points, and the above process is repeated, until all feature points on the target line are in the feature point set, at this time, a plurality of feature point sets can be obtained. Taking the feature point set a as an example, the to-be-distinguished region corresponding to the feature point set a should be the line segment of the feature point 7 and the feature point 14 on the target line, the line segment between the feature point 7 and the corresponding reference point 7', the line segment between the feature point 14 and the corresponding reference point 14', and the line segment of the reference point 7' and the reference point 14' on the edge line, the four line segments enclose a to-be-distinguished region. Please refer to Figure 5 Fig. 2 shows a to-be-distinguished region division schematic diagram in an embodiment of the present application.

[0093] At this point, each blood vessel in the blood vessel region in each CT image has been divided, and the to-be-distinguished region has been obtained, and the shape feature of the to-be-distinguished region and the area change between the to-be-distinguished region and the corresponding region in other CT images can be continuously analyzed, so as to determine the blood vessel overlap representation evaluation index of each to-be-distinguished region.

[0094] Preferably, in one embodiment of the present application, the method for obtaining the blood vessel overlap representation evaluation index comprises:

[0095] Matching the feature points in different CT images: Since the acquisition process of CT images follows certain scanning intervals, scanning directions and other parameters, there is a certain spatial and temporal relationship between different CT images. Therefore, the relative movement distance between the feature points in each CT image and the pixel points in the adjacent CT image can be calculated, so as to determine the corresponding positions of the feature points in each to-be-distinguished region in each CT image in other CT images, and then the corresponding regions can be obtained, and the corresponding regions of each to-be-distinguished region in each CT image in other CT images are taken as the contrast regions.

[0096] Since the area of the region where the blood vessel overlap occurs may present different sizes in different CT images, and the area of the region where the blood vessel overlap does not occur does not change significantly, in each to-be-distinguished region and all the corresponding contrast regions, the difference between the maximum area and the minimum area is taken as the first overlap representation evaluation factor corresponding to each to-be-distinguished region. The larger the first overlap representation evaluation factor is, the more significant the area change is, and the larger the blood vessel overlap representation evaluation index is, and the more likely it is a blood vessel overlap region.

[0097] Then, the shape feature of each to-be-distinguished region is analyzed. When the blood vessel overlap occurs, certain undulations occur, which can be represented by the value of the curvature. Therefore, in each to-be-distinguished region, the midpoint of the line segment formed by each feature point and the corresponding reference point is obtained, and the line connecting all the midpoints is taken as the center line. Here, the least square method is used to fit to obtain the center line. The more undulations appear on the center line, the more likely the to-be-distinguished region is a blood vessel overlap region.

[0098] The mean value of the curvature values of all the midpoints on each center line is calculated as the bending degree value, and the variance of the curvature values of all the midpoints on each center line is taken as the fluctuation degree value. The larger the bending degree value and the fluctuation degree value are, the more undulations and changes in the shape of the to-be-distinguished region occur. Therefore, the product of the normalized values of the bending degree value and the fluctuation degree value corresponding to each to-be-distinguished region is taken as the second overlap representation evaluation factor corresponding to each to-be-distinguished region. The second overlap representation evaluation factor comprehensively considers the bending degree and the shape stability of the to-be-distinguished region. The larger the second overlap representation evaluation factor is, the more likely the to-be-distinguished region is a blood vessel overlap region.

[0099] Finally, the sum of the first overlap representation evaluation factor and the second overlap representation evaluation factor corresponding to each to-be-distinguished region is normalized, and the normalized value is taken as the blood vessel overlap representation evaluation index of each to-be-distinguished region. At this time, the greater the blood vessel overlap representation evaluation index, the more significant the area change of the to-be-distinguished region in different CT images, and the greater the change in the shape of the to-be-distinguished region, so the to-be-distinguished region is more likely to be a blood vessel overlap region. The normalization is a technique known to those skilled in the art, and the selection of the normalization function can be linear normalization or standard normalization. The specific normalization method is not limited here.

[0100] It should be noted that the use of the least squares method is a known technique, and the specific process is not repeated here.

[0101] Based on the foregoing steps, the blood vessel overlap representation evaluation index of each to-be-distinguished region in each CT image can be obtained, and thus it can be determined whether the to-be-distinguished region belongs to a blood vessel overlap region or a non-blood vessel overlap region based on the index.

[0102] Preferably, in an embodiment of the present application, distinguishing the blood vessel overlap region and the non-blood vessel overlap region comprises:

[0103] Based on the foregoing analysis, the greater the blood vessel overlap representation evaluation index, the more likely the to-be-distinguished region is a blood vessel overlap region. Therefore, a preset overlap threshold is set. When the blood vessel overlap representation evaluation index of a to-be-distinguished region is greater than the preset overlap threshold, the to-be-distinguished region is considered to be a blood vessel overlap region.

[0104] When the blood vessel overlap representation evaluation index of a to-be-distinguished region is less than or equal to the preset overlap threshold, the to-be-distinguished region is considered to be a non-blood vessel overlap region.

[0105] It should be noted that in this embodiment of the present application, the preset overlap threshold is 0.6, and the specific value can be adjusted according to the implementation scenario, which is not limited here.

[0106] At this point, the blood vessel overlap region and the non-blood vessel overlap region in the blood vessel region of each CT image can be determined.

[0107] Step S3: In each CT image, each blood vessel overlap region and non-blood vessel overlap region are combined to obtain a plurality of region combinations. In each region combination, the shape similarity between the blood vessel overlap region and the non-blood vessel overlap region and the similarity between the blood vessel overlap representation evaluation indexes of the non-blood vessel overlap regions are analyzed to determine the smooth continuation index of the region combination. The differences between the smooth continuation indexes of all region combinations are analyzed, and the proportion of the blood vessel overlap region in the region combination is combined, so as to screen a matching combination from all region combinations.

[0108] When the blood vessels overlap, the navigation path cannot be clearly presented as expected, that is, it is difficult to determine the matching between the overlapping part and the blood vessels on both sides which do not overlap. Therefore, it is necessary to distinguish which non-overlapping part belongs to the same blood vessel with the overlapping part.

[0109] In the embodiment of the present application, first, in each CT image, for each blood vessel overlapping area, the blood vessel overlapping area is combined with the non-blood vessel overlapping area to obtain a plurality of area combinations. Since the blood vessel edges between the overlapping part and the non-overlapping part of the same blood vessel usually exhibit more smooth connection, and the extension direction of the blood vessel has consistency, specifically, that is, in each area combination, the shape similarity between the blood vessel overlapping area and the non-blood vessel overlapping area should be higher; at the same time, the connection of the same blood vessel at the overlapping area can have higher consistency, so in each area combination, the difference between the blood vessel overlapping area and the non-blood vessel overlapping area is also considered, so as to calculate the smooth continuation index of each area combination, which can represent the possibility of each part in the area combination belonging to the same blood vessel. Further, it is necessary to screen the matching combination with higher matching degree from all the area combinations. In the screening process, since the larger the proportion of the blood vessel overlapping area is, the smaller the matching difficulty is, and the accuracy is improved, therefore, the difference between the smooth continuation indexes of all the area combinations is analyzed, and the proportion of the blood vessel overlapping area in the area combination is taken as a weight, so as to screen the matching combination from all the area combinations.

[0110] First, in each CT image, the area combination is determined. Preferably, in an embodiment of the present application, the method for obtaining the area combination comprises:

[0111] In each CT image, for any blood vessel overlapping area, the non-blood vessel overlapping area connected with the blood vessel overlapping area is taken as a candidate area.

[0112] In two directions of the blood vessel edge line of the blood vessel overlapping area, each optional candidate area and the blood vessel overlapping area form an area combination to obtain all non-repeated area combinations.

[0113] The method for obtaining the area combination is illustrated as follows: if there are four non-blood vessel overlapping areas connected with a blood vessel overlapping area (1), two on each side, which are marked as (2), (3), (4) and (5), please refer to Figure 6Fig. 2 shows a local diagram of the blood vessel overlap region in one embodiment of the present application; (2), (3), (4), (5) are taken as the candidate regions of the blood vessel overlap region, then all the non-repeated region combinations should be {(1), (2), (4)}, {(1), (2), (5)}, {(1), (3), (4)}, {(1), (3), (5)}.

[0114] At this point, the region combination corresponding to each blood vessel overlap region in each CT image can be obtained, and then in each region combination, the shape similarity between the blood vessel overlap region and the non-blood vessel overlap region and the similarity of the blood vessel overlap representation evaluation indexes of the non-blood vessel overlap region are analyzed to determine the smooth continuation index of the region combination.

[0115] Preferably, the method for obtaining the smooth continuation index in one embodiment of the present application comprises:

[0116] Since the blood vessel edges between the overlap and non-overlap parts of the same blood vessel usually exhibit a smoother connection, and the extension direction of the blood vessel has consistency, in this embodiment of the present application, the consistency mainly manifests as the smooth continuation of the center lines of the two edge lines of the blood vessel, and the center line can be taken as the main axis of the blood vessel, which can well reflect the extension and shape features of the blood vessel.

[0117] Therefore, in each region combination, the intersection point between the straight line where the center line of each non-blood vessel overlap region is located and the straight line where the center line of the blood vessel overlap region is located is taken as the intersection point, and the intersection point is the connection position of the main axes of different regions in the region combination, which is helpful to identify the branching, connection and other situations of the blood vessel. The method for obtaining the center line is described in step S2, and at this time each non-blood vessel region corresponds to an intersection point.

[0118] The center line is fitted by all midpoints, and the curvature value at each midpoint can be obtained to reflect the shape feature, thereby helping to analyze the smooth continuity: the Euclidean distance between the midpoint at the position of the maximum curvature value on the center line corresponding to each non-vascular overlapping region and the corresponding intersection point is taken as a position deviation factor; the position of the maximum curvature value represents the position with the maximum bending degree of the non-vascular overlapping region, and therefore if the position deviation factor is smaller, it means that the non-vascular overlapping region and the vascular overlapping region can still maintain good smooth connection at the position with the maximum bending degree; on the contrary, if the position deviation factor is larger, it means that the smooth continuity of the two regions at the intersection is poor. The sum of the position deviation factors corresponding to the two non-vascular overlapping regions in each region combination is negatively correlated to obtain the logical relationship correction, thereby obtaining a smooth factor. The larger the smooth factor is, the better the smooth continuity between the three regions in the region combination is, and it is more likely that they belong to the same blood vessel. The negative correlation mapping here can use the formula exp(-x), where exp() represents the exponential function with the natural constant e as the base, and x represents the independent variable.

[0119] If the three regions in a certain region combination belong to the same blood vessel, the similarity degree between the vascular overlapping representation evaluation indexes of the non-vascular overlapping regions should be higher, reflecting that the connection between the three regions at the vascular overlapping region has a higher consistency. Therefore, in each region combination, the absolute value of the difference between the vascular overlapping representation evaluation indexes of the two non-vascular overlapping regions is taken, the smaller the absolute value is, the higher the similarity degree is, so the absolute value is negatively correlated and normalized to obtain the logical relationship correction, thereby obtaining a similarity factor. The larger the similarity factor is, the higher the consistency between the three regions is, that is, the higher the smooth continuity is. The negative correlation mapping and normalization here can use the formula exp(-x), where exp() represents the exponential function with the natural constant e as the base, and x represents the independent variable.

[0120] Based on the foregoing analysis, the larger the smooth factor and the similarity factor of each region combination are, the better the continuity between the three regions in the region combination is, so the sum of the normalized values of the smooth factor and the similarity factor corresponding to each region combination is taken as the smooth continuity index of each region combination. At this time, the larger the smooth continuity index of a certain region combination is, the more likely it is that the three regions in the region combination belong to the same blood vessel, and the higher the matching degree is. The normalization is a technology known to those skilled in the art, and the selection of the normalization function can be linear normalization or standard normalization, and the specific normalization method is not limited here.

[0121] Further, based on the foregoing process, each blood vessel overlap region can correspond to multiple region combinations, but not all region combinations can correctly represent the matching condition between regions, so it is necessary to screen the region combination with higher matching degree from all region combinations as the matching combination.

[0122] If the three regions in a region combination belong to the same blood vessel, the smooth continuation index of the region combination should be more prominent than that of other region combinations. In addition, the larger the proportion of the blood vessel overlap region, the smaller the proportion of the non-blood vessel overlap region, so the matching difficulty is reduced and the matching accuracy is higher. Therefore, the proportion of the blood vessel overlap region in the region combination is used as a weight to screen the matching combination from all region combinations.

[0123] Preferably, in an embodiment of the present application, the method for obtaining the matching combination comprises:

[0124] In each CT image, for any blood vessel overlap region, the proportion of the smooth continuation index of each region combination corresponding to the blood vessel overlap region in the smooth continuation index of all region combinations corresponding to the blood vessel overlap region is calculated, that is, the sum of the smooth continuation indexes of all region combinations corresponding to the blood vessel overlap region is taken as the denominator, and the smooth continuation index of each region combination is taken as the numerator, so as to obtain the first matching contribution degree of each region combination. The greater the first matching contribution degree of a certain region combination, the more prominent the smooth continuation of the region combination in all region combinations, and thus the higher the matching degree.

[0125] Then, in each region combination corresponding to the blood vessel overlap region, the length average of the two edge lines of the blood vessel in the blood vessel overlap region is calculated as a first length factor, and the length average of the two edge lines of the blood vessel in each non-blood vessel overlap region is calculated as a second length factor; the proportion of the first length factor in the sum of the first length factor and all second length factors is calculated, that is, the sum of the length factors corresponding to the three regions in the region combination (the sum of the first length factor and the two second length factors) is taken as the denominator, and the first length factor is taken as the numerator, to obtain the second matching contribution degree of each region combination. The greater the second matching contribution degree of a certain region combination, the greater the proportion of the blood vessel overlap region in the region combination, and thus the smaller the proportion of the non-blood vessel overlap region, so the matching difficulty is lower and the matching accuracy is higher.

[0126] Finally, the sum of the first matching contribution and the second matching contribution of each region combination corresponding to the blood vessel overlapping area is normalized as a matching coefficient of each region combination. Based on the foregoing analysis, the greater the matching coefficient, the higher the matching degree between the three regions in the region combination, and the more likely the region combination belongs to the same blood vessel. Therefore, among all the region combinations of the blood vessel overlapping area, the region combination with a matching coefficient greater than a preset matching threshold is regarded as a matching combination, and each region in the matching combination is more likely to represent each part of the same blood vessel. The normalization is a technical means known to those skilled in the art, and the normalization function can be linear normalization or standard normalization, and the specific normalization method is not limited herein.

[0127] It should be noted that in this embodiment of the present application, the preset matching threshold is 0.7, and the specific value can be adjusted according to the actual situation, which is not limited herein.

[0128] Step S4: In the matching combination corresponding to each blood vessel overlapping area in each CT image, the difference in gray value between the non-blood vessel overlapping regions is analyzed, the non-blood vessel overlapping regions are hierarchically evaluated, and a hierarchical evaluation result is obtained; and a navigation path of the puncture target point is planned according to the hierarchical evaluation result.

[0129] Based on the foregoing steps, the matching combination corresponding to each blood vessel overlapping area can be obtained, and each part in each matching combination is more likely to belong to the same blood vessel. In this embodiment of the present application, in order to further determine which blood vessels are located in the upper layer and which blood vessels are located in the lower layer, it is considered that the gray values of blood vessels in different layers in the CT image will be different, and therefore the non-blood vessel overlapping regions in the matching combination are hierarchically evaluated by using this feature to obtain a hierarchical evaluation result. The hierarchical evaluation result can more accurately reflect which blood vessels on both sides of the blood vessel overlapping area belong to the same layer, so that when the navigation path of the puncture target point is planned, the other blood vessels can be effectively avoided in the region with blood vessel overlapping, and a more accurate navigation path is obtained.

[0130] First, in the matching combination corresponding to each blood vessel overlapping area in each CT image, the difference in gray value between the non-blood vessel overlapping regions is analyzed, the non-blood vessel overlapping regions are hierarchically evaluated, and a hierarchical evaluation result is obtained.

[0131] Preferably, in an embodiment of the present application, the method for obtaining the hierarchical evaluation result comprises:

[0132] The purpose of this process is to analyze the depth difference represented by the non-blood vessel overlapping regions on both sides of the blood vessel overlapping area, so as to perform hierarchical evaluation, identify the non-blood vessel overlapping regions of the same layer, and achieve a more accurate matching effect.

[0133] Therefore, in each blood vessel overlapping area corresponding to all matching combinations, in all non-blood vessel overlapping areas along each direction of the blood vessel overlapping area, the average value of the gray scale of all pixel points in each non-blood vessel overlapping area is calculated as the gray scale characteristic value of each non-blood vessel overlapping area.

[0134] Because the blood vessels located in the upper layer are closer to the scanning device, the gray scale value shown in the CT image is larger, and vice versa, the blood vessels located in the lower layer are farther away from the scanning device, the gray scale value shown in the CT image is smaller, so all non-blood vessel overlapping areas in each direction are arranged in ascending order according to the gray scale characteristic value to obtain a sorting sequence.

[0135] Finally, in the two sorting sequences corresponding to the blood vessel overlapping area, the two non-blood vessel overlapping areas with the same serial number value belong to the blood vessels of the same level.

[0136] After the non-blood vessel overlapping areas on both sides of the blood vessel overlapping area are evaluated in terms of the level, the relative position of the blood vessel can be understood more deeply, so that the penetration of other blood vessels can be effectively avoided, and before entering the blood vessel overlapping area, the planning path of the retrievable inferior vena cava filter placement puncture target at the non-blood vessel overlapping area has been obtained, then the puncture target in the non-blood vessel overlapping area corresponding to the existing planning path can be selected, and then the navigation path planning of the retrievable inferior vena cava filter placement puncture target in the presence of blood vessel overlapping can be completed. Figure 7 Fig. 4 shows the navigation path planning process in one embodiment of the present application.

[0137] To sum up, when planning the navigation path of the inferior vena cava filter puncture target point, the influence of the blood vessel overlap may exist, which affects the planning accuracy of the navigation path, so in the embodiment of the application, the influence of the blood vessel overlap is mainly analyzed. First, a plurality of CT images of the inferior vena cava of the patient are acquired, and the blood vessel region is segmented. Since the shape of the blood vessel changes when the blood vessel overlaps, that is, the distance between the edge lines of the blood vessel changes obviously, the blood vessel region is preliminarily divided based on this feature, and the region to be divided is obtained. Further, when the blood vessels overlap, the area at the overlapping position may present different changes in different CT images, so the shape feature of each region to be analyzed and the area change of the region to be analyzed and the corresponding region in other CT images are comprehensively considered to determine the blood vessel overlap representation evaluation index of the region to be divided, and the blood vessel overlap region and the non-blood vessel overlap region are divided based on the index. In order to determine which non-blood vessel overlap regions on both sides belong to the same blood vessel as the blood vessel overlap region, the blood vessel overlap region and the non-blood vessel overlap region are combined in the embodiment of the application to obtain a plurality of region combinations. Since if they belong to the same blood vessel, the spatial continuity and shape should be relatively consistent and smooth, so in each region combination, the shape similarity between the blood vessel overlap region and the non-blood vessel overlap region and the similarity between the overlap representation evaluation indexes of the non-blood vessel overlap regions are analyzed to determine the smooth continuation index of the region combination; since the larger the proportion of the blood vessel overlap region, the smaller the proportion of the non-blood vessel overlap part, the matching difficulty will be reduced, and the weight should be increased, so in all region combinations, based on the difference between the smooth continuation indexes and the proportion of the blood vessel overlap region, a matching combination is screened out, and the regions in each matching combination represent that they are more likely to belong to the same blood vessel. Further, since the gray values of the blood vessels at different levels in the CT images will be different, in the application, for each matching combination corresponding to the blood vessel overlap region, the difference of the gray values between the non-blood vessel overlap regions is analyzed, so that the level of the non-blood vessel overlap region is evaluated, and a level evaluation result is obtained. The level evaluation result can more accurately represent which regions belong to the same blood vessel, and then for the region with blood vessel overlap, the navigation path of the puncture target point can be more accurately planned according to the level evaluation result.

[0138] It should be noted that the above-mentioned embodiment of the application is only for description, and does not represent the advantages and disadvantages of the embodiment. The processes depicted in the 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.

[0139] The various embodiments described in this specification are presented by way of example, and each embodiment is not inherently more important than any other embodiment.

Claims

1. An auxiliary path planning method for an inferior vena cava filter implantation puncture target point, characterized in that, The method comprises: acquiring a plurality of CT images of the inferior vena cava of a patient, and segmenting a blood vessel region in each CT image; in each CT image, based on the distance variation between the two edge lines of each blood vessel in the blood vessel region, the to-be-differentiated region is divided; the shape feature of each to-be-differentiated region in each CT image and the area variation between each to-be-differentiated region and the corresponding region in other CT images are analyzed, and a blood vessel overlap representation evaluation index of each to-be-differentiated region is determined, so as to differentiate the blood vessel overlap region and the non-blood vessel overlap region; the blood vessel overlap representation evaluation index is used to reflect whether the blood vessels overlap; in each CT image, each blood vessel overlap region and non-blood vessel overlap region are combined to obtain a plurality of region combinations; in each region combination, the shape similarity between the blood vessel overlap region and the non-blood vessel overlap region is analyzed, and the similarity between the blood vessel overlap representation evaluation indexes of the non-blood vessel overlap regions is analyzed, so as to determine a smooth continuation index of the region combination; the difference between the smooth continuation indexes of all region combinations is analyzed, and the proportion of the blood vessel overlap region in the region combination is combined, so as to screen a matching combination from all region combinations; the smooth continuation index is used to represent the possibility that each part in the region combination belongs to the same blood vessel; in the matching combination corresponding to each blood vessel overlap region in each CT image, the difference of the gray value between the non-blood vessel overlap regions is analyzed, the hierarchical evaluation of the non-blood vessel overlap regions is carried out, and a hierarchical evaluation result is obtained; the hierarchical evaluation result is used to represent which regions belong to the same blood vessel; and a navigation path of a puncture target point is planned according to the hierarchical evaluation result.

2. The auxiliary path planning method for an inferior vena cava filter implantation puncture target point according to claim 1, characterized in that, The method for obtaining the to-be-differentiated region comprises: in each CT image, any one of the two edge lines of each blood vessel is selected as a target line; on the target line, a feature point is obtained based on a SURF feature point detection algorithm, a straight line perpendicular to the target line is drawn through each feature point, and the intersection of the straight line and the other edge line is taken as a reference point corresponding to each feature point; the distance between each feature point and the corresponding reference point is taken as a distance factor; in the initial state, all feature points are not in the feature point set on the target line, a feature point is randomly selected as a to-be-tested point, and the to-be-tested point corresponds to a feature point set; taking the to-be-tested point as a starting point, the feature points on both sides of the target line are traversed respectively, for each traversed feature point, the difference of the distance factors between the feature point and the feature points in the feature point set is analyzed, a merging index is obtained, when the merging index meets a preset condition, the traversed feature point is merged into the feature point set of the to-be-tested point, and the traversal is continued; if not, the traversal is stopped; wherein the preset condition is set as the merging index being greater than or equal to a preset merging threshold; on the target line, a feature point not existing in the feature point set is randomly selected as a new to-be-tested point, and the above traversal process is repeated until all feature points exist in the feature point set, and all feature point sets on the target line are obtained. In each feature point set, the two feature points farthest from each other on the target line are taken as end points, and the line segment between the end points and the corresponding reference points, the line segment of the two end points on the edge line, and the line segment of the two end points on the target line form a region to be distinguished.

3. The auxiliary path planning method for an inferior vena cava filter implantation puncture target point according to claim 2, characterized in that, The method for obtaining the merging index comprises: taking each traversed feature point as a point to be analyzed; calculating the absolute value of the difference between the distance factor of the point to be analyzed and the distance factor of each feature point in the feature point set corresponding to the point to be measured as a difference factor; performing negative correlation mapping on the mean value of all the difference factors corresponding to the point to be analyzed and normalizing the value as the merging index.

4. The auxiliary path planning method for an inferior vena cava filter implantation puncture target point according to claim 2, characterized in that, The method for obtaining the blood vessel overlap representation evaluation index comprises: matching the feature points in different CT images to determine the corresponding regions of each region to be distinguished in each CT image in other CT images, and taking each region to be distinguished in each CT image as a contrast region in other CT images; in each region to be distinguished and all the corresponding contrast regions, taking the difference between the maximum region area and the minimum region area as the first overlap representation evaluation factor corresponding to each region to be distinguished; in each region to be distinguished, taking the line connecting the midpoint of the line segment between the feature point and the corresponding reference point as the center line, calculating the mean value of the curvature values at all the midpoints on each center line as the bending degree value, and taking the variance of the curvature values at all the midpoints on each center line as the fluctuation degree value; normalizing the product of the bending degree value and the fluctuation degree value corresponding to each region to be distinguished as the second overlap representation evaluation factor corresponding to each region to be distinguished; normalizing the sum of the first overlap representation evaluation factor and the second overlap representation evaluation factor corresponding to each region to be distinguished as the blood vessel overlap representation evaluation index of each region to be distinguished.

5. The auxiliary path planning method for an inferior vena cava filter implantation puncture target point according to claim 1, characterized in that, The method for distinguishing the blood vessel overlap region and the non-blood vessel overlap region comprises: when the blood vessel overlap representation evaluation index of a certain region to be distinguished is greater than a preset overlap threshold, the region to be distinguished is considered as a blood vessel overlap region; when the blood vessel overlap representation evaluation index of a certain region to be distinguished is less than or equal to the preset overlap threshold, the region to be distinguished is considered as a non-blood vessel overlap region.

6. The auxiliary path planning method for an inferior vena cava filter implantation puncture target point according to claim 4, characterized in that, The method for obtaining the region combination comprises: in each CT image, for any one blood vessel overlap region, taking the non-blood vessel overlap region connected to the blood vessel overlap region as a candidate region; in two directions along the blood vessel edge line of the blood vessel overlap region, each candidate region is optionally combined with the blood vessel overlap region to form a region combination, and all non-repeated region combinations are obtained.

7. The auxiliary path planning method for an inferior vena cava filter implantation puncture target point according to claim 6, characterized in that, The method for obtaining the smooth continuation index comprises: in each region combination, taking the intersection point between the straight line where the center line of each non-blood vessel overlap region is located and the straight line where the center line of the blood vessel overlap region is located as a convergence point; taking the Euclidean distance between the midpoint of the maximum curvature value on the center line corresponding to each non-blood vessel overlap region and the corresponding convergence point as a position deviation factor, and performing negative correlation mapping on the sum of the position deviation factors corresponding to two non-blood vessel overlap regions to obtain the smooth factor. The absolute value of the difference between the two non-vascular overlap region vascular overlap characterization evaluation indexes is negatively correlated and normalized, and the value after normalization is taken as a similarity factor; The sum of the corresponding smooth factor and the similarity factor of each region combination is normalized, and the value after normalization is taken as a smooth continuation index of each region combination.

8. The auxiliary path planning method for an inferior vena cava filter implantation puncture target point according to claim 1, characterized in that, The acquisition method of the matching combination includes: In each CT image, for any one vascular overlap region, the proportion of the smooth continuation index of each region combination corresponding to the vascular overlap region in the smooth continuation index of all region combinations corresponding to the vascular overlap region is taken as a first matching contribution degree; In each region combination corresponding to the vascular overlap region, the length average of the two edge lines of the blood vessels in the vascular overlap region is calculated as a first length factor, and the length average of the two edge lines of the blood vessels in each non-vascular overlap region is calculated as a second length factor, and the proportion of the first length factor in the sum of the first length factor and all second length factors is taken as a second matching contribution degree; The sum of the first matching contribution degree and the second matching contribution degree of each region combination corresponding to the vascular overlap region is normalized, and the value after normalization is taken as a matching coefficient; In all region combinations corresponding to the vascular overlap region, the region combination with a matching coefficient greater than a preset matching threshold is taken as a matching combination.

9. The auxiliary path planning method for an inferior vena cava filter implantation puncture target point according to claim 8, characterized in that, The acquisition method of the hierarchical evaluation result includes: In all matching combinations, in all non-vascular overlap regions along each direction of the vascular overlap region, the average of the gray values of all pixel points in each non-vascular overlap region is calculated as a gray feature value of each non-vascular overlap region, and all non-vascular overlap regions are arranged in ascending order according to the gray feature value to obtain a sorting sequence; In the two sorting sequences corresponding to the vascular overlap region, two non-vascular overlap regions with the same sequence number belong to the same level of blood vessels.

10. The auxiliary path planning method for an inferior vena cava filter implantation puncture target point according to claim 1, characterized in that, The acquisition method of the blood vessel region includes: Each CT image is taken as the input of the pre-trained neural network, so as to output the blood vessel region in each CT image.

Images