Three-dimensional navigation method, device, equipment and medium for vascular interventional surgery robot

Abstract

The present invention provides a three-dimensional navigation method, device, equipment and medium for a vascular interventional surgery robot. The method comprises: determining a puncture starting point, a puncture end point and a target node set in a three-dimensional vascular model, traversing the target node set to obtain a plurality of delivery trajectories; performing node screening on each delivery trajectory based on the curvature value of each node on each delivery trajectory to obtain a target number of feature nodes; screening a plurality of terminal points based on the target number of feature nodes on each delivery trajectory, and screening a target terminal point set from the plurality of terminal points; extracting a vascular centerline, determining a centerline node that matches each target terminal point in the target terminal point set; performing path planning on the target terminal point set based on an improved A-Star algorithm to obtain an optimal centerline path, thereby traversing all trajectory postures and performing trajectory screening during traversal to reduce the amount of calculation, and using the improved A-Star algorithm to quickly and accurately perform path planning.

Description

Technical Field

[0001] The present invention relates to the field of vascular interventional surgery, and in particular to a three-dimensional navigation method, device, equipment and medium for a vascular interventional surgery robot. Background Art

[0002] Cardiovascular disease is the biggest threat to human health in the world. Vascular interventional surgery has gradually been used to treat cardiovascular and cerebrovascular diseases due to its advantages such as small incision and fast recovery. Compared with manual delivery by interventional radiologists, vascular robots have the advantages of safety, efficiency and accuracy. They can assist vascular interventional physicians in quantitatively controlling the axial delivery, withdrawal and rotation of interventional surgical instruments during interventional surgery.

[0003] At present, when using vascular robots to assist vascular interventional physicians in performing interventional surgeries, three-dimensional intravascular navigation is usually required to help vascular interventional physicians deliver surgical instruments without relying on X-ray angiography. Three-dimensional intravascular navigation is a navigation algorithm for path planning in a three-dimensional vascular model. This task requires finding a limited path trajectory in a complex pipeline obstacle. When performing three-dimensional intravascular navigation, it is also necessary to simulate the elastic rod-type interventional surgical instrument for three-dimensional interventional surgical instrument. Traditional three-dimensional interventional surgical instrument simulation systems use a single posture of the surgical instrument to simulate the path trajectory in a static environment. However, using a single posture to simulate the path trajectory will lead to too limited simulation results. When simulating a single posture of the surgical instrument, the surgical instrument will be divided into several equally spaced control points, and a three-dimensional frame will be established at each node, and then the posture will be calculated for all control points. However, in the actual delivery process, most of the internal stresses of the surgical instrument are small values. Repeated calculation of low internal stress points will increase a large amount of invalid calculations, which will cause the path planning to take too long. Summary of the invention

[0004] The present invention provides a three-dimensional navigation method, device, equipment and medium for a vascular interventional surgery robot, which are used to solve the defects in the prior art that the path planning results are too limited and time-consuming, and to achieve faster and more accurate path planning.

[0005] The present invention provides a three-dimensional navigation method for a vascular interventional surgery robot, comprising:

[0006] Determine the puncture starting point, puncture end point and target node set in the three-dimensional blood vessel model, and traverse the target node set to obtain a plurality of delivery trajectories, wherein the delivery trajectory is a trajectory starting from the puncture starting point, passing through the target node and finally reaching the puncture end point;

[0007] Based on the curvature value of each node on each delivery trajectory, each delivery trajectory is screened to obtain a target number of feature nodes;

[0008] Filtering a plurality of terminal points from a plurality of delivery trajectories based on a target number of characteristic nodes on each delivery trajectory, and filtering a target terminal point set from the plurality of terminal points;

[0009] Extracting a blood vessel centerline in the three-dimensional blood vessel model, and determining a centerline node matching each target terminal point in the target terminal point set, wherein the centerline node is a node located on the blood vessel centerline;

[0010] Performing path planning on the target terminal point set based on the improved A-Star algorithm to obtain an optimal centerline path, and performing three-dimensional navigation according to the optimal centerline path;

[0011] The total cost calculation formula of the improved A-Star algorithm is as follows: f(n)=g(n)+h′(n); wherein f(n) represents the total cost of the target terminal point n, g(n) represents the cost of the target terminal point n from the puncture starting point, and h′(n) represents the improved Euclidean distance between the target terminal point n and the puncture end point; the cost calculation formula of h′(n) is as follows: Among them, the Indicates the distance between the target terminal point n and the center line node se i The Euclidean distance, Indicates the centerline node se i The centerline distance from the puncture endpoint.

[0012] According to a three-dimensional navigation method for a vascular interventional surgery robot provided by the present invention, the step of determining the centerline node that matches each target terminal point in the target terminal point set specifically includes:

[0013] Selecting a plurality of centerline nodes from the centerline of the blood vessel;

[0014] Selecting a set of centerline nodes located at the target position of each target terminal point from the plurality of centerline nodes respectively;

[0015] Calculate the node distances between each centerline node in the centerline node set of each target terminal point and each target terminal point respectively;

[0016] The centerline nodes in the centerline node set of each target terminal point are sorted from small to large based on the node distance, and the centerline nodes in the preset ranking are used as the centerline nodes matched by each target terminal point.

[0017] According to a three-dimensional navigation method for a vascular interventional surgery robot provided by the present invention, path planning is performed on the target terminal point set based on the improved A-Star algorithm to obtain the optimal centerline path, specifically including:

[0018] Calculate the total cost of each target terminal point in the target terminal point set based on the improved A-Star algorithm;

[0019] A target terminal point with the minimum total cost is determined, and an optimal centerline path is planned based on the target terminal point with the minimum total cost.

[0020] According to a three-dimensional navigation method for a vascular interventional surgery robot provided by the present invention, the node screening of each delivery trajectory is performed based on the curvature value of each node on each delivery trajectory to obtain a target number of characteristic nodes, specifically including:

[0021] Filter out a number of target nodes on each delivery trajectory whose curvature value is greater than or equal to a preset curvature value;

[0022] Determining a target number of characteristic nodes on each delivery trajectory according to the delivery displacement length of the surgical robot;

[0023] Sort several target nodes on each delivery trajectory from large to small based on the curvature value;

[0024] The target nodes ranked in the front on each delivery trajectory are respectively used as feature nodes on each delivery trajectory, and the target number of feature nodes are obtained.

[0025] According to a three-dimensional navigation method for a vascular interventional surgery robot provided by the present invention, the method of selecting a plurality of terminal points from a plurality of delivery trajectories based on a target number of characteristic nodes on each delivery trajectory, and selecting a target terminal point set from the plurality of terminal points, specifically includes:

[0026] Based on the target number of characteristic nodes on each delivery trajectory, each delivery trajectory is divided into a plurality of elastic rod trajectories, and a total elastic potential energy value of each delivery trajectory is calculated based on the plurality of elastic rod trajectories;

[0027] Selecting at least one effective delivery trajectory from the plurality of delivery trajectories based on the total elastic potential energy value, and selecting a plurality of terminal points from the at least one effective delivery trajectory;

[0028] Clustering the plurality of terminal points into at least one type of terminal point set based on a clustering algorithm;

[0029] A target terminal point set is screened out from the at least one type of terminal point set according to the weighted average position of each type of terminal point set.

[0030] According to a three-dimensional navigation method for a vascular interventional surgery robot provided by the present invention, the total elastic potential energy value of each delivery trajectory is calculated based on a plurality of elastic rod trajectories, specifically including:

[0031] The bending potential energy of each segment of the elastic rod trajectory is calculated based on the pseudo-rigid body model;

[0032] The torsional potential energy of each elastic rod trajectory is calculated based on Kirchhoff's elastic rod theory;

[0033] The total elastic potential energy value of each delivery trajectory is calculated according to the bending potential energy and the torsional potential energy of each segment of the elastic rod trajectory.

[0034] According to a three-dimensional navigation method for a vascular interventional surgery robot provided by the present invention, before performing path planning on the target terminal point set based on the improved A-Star algorithm to obtain the optimal centerline path, the method further includes:

[0035] Determining the radius of the optimized spherical model according to the delivery displacement length of the surgical robot;

[0036] constructing an optimized spherical model at the target terminal point set based on the radius of the optimized spherical model;

[0037] The target terminal point set is optimized based on the optimized spherical model.

[0038] The present invention also provides a three-dimensional navigation device for a vascular interventional surgery robot, comprising:

[0039] A traversal unit, used to determine the puncture starting point, the puncture end point and the target node set in the three-dimensional blood vessel model, and traverse the target node set to obtain a plurality of delivery trajectories, wherein the delivery trajectory is a trajectory starting from the puncture starting point, passing through the target node and finally reaching the puncture end point;

[0040] A first screening unit is used to screen nodes on each delivery trajectory based on the curvature value of each node on each delivery trajectory to obtain a target number of feature nodes;

[0041] A second screening unit is used to screen out a plurality of terminal points from the plurality of delivery trajectories based on a target number of characteristic nodes on each delivery trajectory, and screen out a target terminal point set from the plurality of terminal points;

[0042] An extraction unit, configured to extract a blood vessel centerline from the three-dimensional blood vessel model, and determine a centerline node matching each target terminal point in the target terminal point set, wherein the centerline node is a node located on the blood vessel centerline;

[0043] A planning unit is used to perform path planning for the target terminal point set based on the improved A-Star algorithm to obtain an optimal centerline path, and perform three-dimensional navigation according to the optimal centerline path; the total cost calculation formula of the improved A-Star algorithm is as follows: f(n)=g(n)+h′(n); wherein f(n) represents the total cost of the target terminal point n, g(n) represents the cost of the target terminal point n from the puncture starting point, and h′(n) represents the improved Euclidean distance of the target terminal point n from the puncture end point; the cost calculation formula of h′(n) is as follows: Among them, the Indicates the distance between the target terminal point n and the center line node se i The Euclidean distance, Indicates the centerline node se i The centerline distance from the puncture endpoint.

[0044] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the three-dimensional navigation method of the vascular interventional surgery robot as described above is implemented.

[0045] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described three-dimensional navigation methods for a vascular interventional surgery robot.

[0046] The present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-mentioned three-dimensional navigation methods for vascular interventional surgery robots.

[0047] The present invention provides a three-dimensional navigation method, device, equipment and medium for a vascular interventional surgery robot. The method determines the puncture starting point, puncture end point and target node set in a three-dimensional vascular model, and traverses the target node set to obtain a plurality of delivery trajectories. Then, based on the curvature value of each node on each delivery trajectory, each delivery trajectory is node-screened to obtain a target number of characteristic nodes. Then, based on the target number of characteristic nodes on each delivery trajectory, a plurality of terminal points are screened from the plurality of delivery trajectories, and a target terminal point set is screened from the plurality of terminal points. Finally, the vascular centerline is extracted in the three-dimensional vascular model to determine the target The centerline nodes that match each target terminal point in the terminal point set are used for path planning of the target terminal point set based on the improved A-Star algorithm to obtain the optimal centerline path, and three-dimensional navigation is performed based on the optimal centerline path. Thus, by traversing all possible trajectories, all postures of interventional surgical instruments in the delivery process are simulated, which solves the limitations of simulation results caused by a single posture simulation process in the prior art. When traversing all possible trajectories, multiple trajectory screenings are performed to reduce the amount of calculation in the simulation process, and the improved A-Star algorithm is used to achieve the effect of quickly and accurately obtaining the optimal centerline path. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0049] Figure 1 This is one of the flow diagrams of the three-dimensional navigation method of the vascular intervention surgery robot provided by the present invention;

[0050] Figure 2 This is the second flow chart of the three-dimensional navigation method of the vascular intervention surgery robot provided by the present invention;

[0051] Figure 3 This is the third flow chart of the three-dimensional navigation method of the vascular interventional surgery robot provided by the present invention;

[0052] Figure 4 A schematic diagram of a pseudo-rigid body model of an interventional surgical instrument provided by the present invention;

[0053] Figure 5 This is the fourth flow chart of the three-dimensional navigation method of the vascular intervention surgery robot provided by the present invention;

[0054] Figure 6 It is a structural schematic diagram of the three-dimensional navigation device of the vascular intervention surgery robot provided by the present invention;

[0055] Figure 7 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Combine the following Figure 1-Figure 5 The present invention describes the three-dimensional navigation method of a vascular interventional surgery robot.

[0058] Figure 1 This is one of the flow charts of the three-dimensional navigation method of the vascular interventional surgery robot provided by the present invention, such as Figure 1 As shown, the method includes:

[0059] Step 100, determining the puncture starting point, puncture end point and target node set in the three-dimensional blood vessel model, and traversing the target node set to obtain a plurality of delivery trajectories, wherein the delivery trajectory is a trajectory starting from the puncture starting point, passing through the target node and finally reaching the puncture end point;

[0060] Specifically, the three-dimensional vascular model in the present invention is constructed based on the CTA image of the patient containing the target vascular part. For example, in actual applications, when the patient needs to undergo radial artery puncture and coronary intervention, it is necessary to first obtain the CTA image of the patient containing the radial artery part, and then perform three-dimensional vascular modeling on the CTA image containing the radial artery part to obtain a three-dimensional vascular model. It should be noted that the present invention uses existing modeling technology to perform three-dimensional vascular modeling, which will not be repeated here.

[0061] In the present invention, the puncture starting point refers to the starting point for delivery of the interventional instrument of the vascular interventional surgery robot, and the puncture end point refers to the end point for delivery of the interventional instrument of the vascular interventional surgery robot. The target node set is obtained by node selection from the three-dimensional vascular model. When performing node selection, a target node within a preset step range can be determined first. For example, the pixel point in the three-dimensional vascular model at the preset step position from the puncture starting point is set as the current target node, and then the pixel point in the three-dimensional vascular model at the preset step position from the current target node is set as the target node again, thereby performing node traversal in sequence until the puncture end point is traversed. In the present invention, node traversal can also be performed from the puncture end point to the puncture starting point to obtain the target node set, and there is no restriction on this.

[0062] In addition, in practical applications, there are usually complex vascular morphologies such as U-shaped structures in blood vessels, and in some specific parts, there are even multiple vascular pathways. Therefore, the present invention also sets the pixel points located at the vascular branches in the three-dimensional vascular model as target nodes, thereby ensuring the accuracy of path simulation. In addition, the present invention can also use other methods to obtain the target node set, which is not limited.

[0063] By traversing all nodes in the target node set, the entire delivery trajectory of the interventional instrument of the vascular interventional surgery robot from the puncture starting point to the puncture end point is obtained, and all postures of the interventional instrument during the delivery process are simulated, thereby solving the defect of the existing technology that only a single posture simulation is performed, resulting in the simulation result being too limited.

[0064] Step 200, screening nodes on each delivery trajectory based on the curvature value of each node on each delivery trajectory to obtain a target number of feature nodes;

[0065] Specifically, the present invention adopts K points to calculate the curvature value, wherein the curvature value Cur(i) of the i-th node is calculated as follows:

[0066]

[0067] Among them, S i =(x,y,z), that is, S i Refers to the three-dimensional coordinates of the i-th node, S i-k Refers to the three-dimensional coordinates of the Kth node in the direction from the puncture starting point to the puncture end point, S i-2k Refers to the three-dimensional coordinates of the 2Kth node in the direction from the puncture start point to the puncture end point.

[0068] In the present invention, the characteristic node refers to the node whose curvature value is not less than the minimum curvature value. It should be noted that during the delivery process of the vascular interventional surgical robot, when the interventional instrument is not squeezed by the vascular wall, it will maintain a large range of linear motion under the influence of elastic potential energy. Due to the limited support of the vascular wall, exceeding a certain force will cause damage to the vascular wall. The curvature value that the interventional instrument can be bent under a specific vascular support force is limited. The larger the curvature value, the smoother the corresponding trajectory. Therefore, the present invention adopts a method based on curvature constraints to filter the nodes of the delivery trajectory, thereby reducing the number of calculation nodes and improving calculation efficiency.

[0069] Step 300, screening out a plurality of terminal points from a plurality of delivery trajectories based on a target number of characteristic nodes on each delivery trajectory, and screening out a target terminal point set from the plurality of terminal points;

[0070] Specifically, the delivery trajectory is first reduced to several segments of elastic rod trajectories according to the characteristic nodes on each delivery trajectory, so that the model can calculate the total elastic potential energy value of each delivery trajectory, and then judge the collision type between the delivery trajectory and the inner wall of the blood vessel by the total elastic potential energy value. It is easy to understand that when the total elastic potential energy value of the delivery trajectory is too large, it indicates that the delivery trajectory will be strongly squeezed by the inner wall of the blood vessel. In other words, the collision type between the delivery trajectory and the inner wall of the blood vessel is a strong collision, which will cause damage to the blood vessel wall.

[0071] In the present invention, after the collision type between the delivery trajectory and the inner wall of the blood vessel is determined by the total elastic potential energy value, the delivery trajectory that may cause damage to the blood vessel wall is eliminated, thereby ensuring that the simulation result deviation of the delivery trajectory is reduced.

[0072] Specifically, the terminal point refers to the node on the delivery trajectory within the preset range of the puncture endpoint, and the target terminal point set refers to the terminal point set closest to the puncture endpoint. In actual applications, after removing the delivery trajectories that may cause damage to the blood vessel wall, several terminal points are screened out from the remaining delivery trajectories, and then the several terminal points are clustered into at least one type of terminal point set, and the terminal point set closest to the puncture endpoint is set as the target terminal point set, thereby obtaining the puncture position with the highest probability in the actual interventional surgery.

[0073] Step 400, extracting a blood vessel centerline in the three-dimensional blood vessel model, and determining a centerline node matching each target terminal point in the target terminal point set, wherein the centerline node is a node located on the blood vessel centerline;

[0074] Specifically, the blood vessel centerline has the same topology and similar shape as the blood vessel, and can well reflect the direction and spatial structure of the blood vessel. Therefore, the present invention uses the extracted blood vessel centerline as a reference for the total cost of global path planning, thereby ensuring that the resulting path planning is more accurate.

[0075] In the present invention, the vascular centerline can be extracted from the three-dimensional vascular model based on the existing known vascular centerline extraction technology, which will not be described in detail here. In practical applications, since the three-dimensional coordinates of each target terminal point are different, the centerline node closest to each target terminal point is also different.

[0076] Step 500, performing path planning on the target terminal point set based on the improved A-Star algorithm to obtain an optimal centerline path, and performing three-dimensional navigation according to the optimal centerline path;

[0077] The total cost calculation formula of the improved A-Star algorithm is as follows: f(n)=g(n)+h′(n); wherein f(n) represents the total cost of the target terminal point n, g(n) represents the cost of the target terminal point n from the puncture starting point, and h′(n) represents the improved Euclidean distance between the target terminal point n and the puncture end point; the cost calculation formula of h′(n) is as follows: Among them, the Indicates the distance between the target terminal point n and the center line node se i The Euclidean distance, Indicates the centerline node se i The centerline distance from the puncture endpoint.

[0078] It should be noted that, during the vascular delivery process, it is necessary to find the position of the instrument end closest to the puncture end point, so as to provide a warning reference for the doctor before the instrument end is delivered to certain specific blood vessels. Currently common path planning algorithms include particle swarm optimization and its derivative algorithms, spatial projection matrix method, fast random tree method, random roadmap method, artificial potential field method, vector field histogram algorithm, etc., but none of them can achieve global optimization. Therefore, the present invention adopts A-Star algorithm for path planning.

[0079] In practical applications, because there are often complex vascular morphologies such as U-shaped structures in the vascular model, and in certain specific locations, there are even multiple vascular pathways. The vascular centerline simplifies the three-dimensional vascular model into a connected trajectory composed of line segments. Therefore, the improved A-Star algorithm is used in the present invention to perform path planning on the target terminal point set to obtain the optimal centerline path.

[0080] Specifically, the improved A-Star algorithm is used to perform path planning on the target terminal point set to obtain the optimal centerline path, which specifically includes:

[0081] The total cost of each target terminal point in the target terminal point set is calculated based on the improved A-Star algorithm; the target terminal point with the minimum total cost is determined, and the optimal centerline path is planned based on the target terminal point with the minimum total cost.

[0082] Therefore, the extracted vascular centerline is used as a reference for the total cost of global path planning, replacing the Euclidean distance between the target terminal point and the puncture end point in the original A-Star algorithm. This makes path planning based on the total cost of the target terminal point more convincing and can obtain an optimal centerline path that is more in line with the actual situation.

[0083] In addition, in another embodiment, before performing path planning on the target terminal point set based on the improved A-Star algorithm to obtain the optimal centerline path, the method further includes:

[0084] The radius of the optimized spherical model is determined according to the delivery displacement length of the surgical robot; the optimized spherical model is constructed at the target terminal point set based on the radius of the optimized spherical model; and the target terminal point set is optimized based on the optimized spherical model.

[0085] Specifically, the optimized spherical model is a model for optimizing the target terminal point set, and the center of the sphere may be located at the weighted average position of the target terminal point set, wherein the spherical radius r of the optimized spherical model is related to the delivery displacement length L of the surgical robot, and the specific relationship is:

[0086] r=c r *L; where c r Is a constant.

[0087] In the present invention, after determining the radius of the optimized spherical model, the weighted average position of the target terminal point set is used as the sphere center to construct the optimized spherical model, and then the terminal points in the target terminal point set that are outside the optimized spherical model are filtered out, thereby improving the accuracy of trajectory planning.

[0088] In the present invention, a puncture starting point, a puncture end point and a target node set are determined in a three-dimensional vascular model, and the target node set is traversed to obtain a plurality of delivery trajectories. Then, based on the curvature value of each node on each delivery trajectory, node screening is performed on each delivery trajectory to obtain a target number of feature nodes. Then, based on the target number of feature nodes on each delivery trajectory, a plurality of terminal points are screened from the plurality of delivery trajectories, and a target terminal point set is screened from the plurality of terminal points. Finally, the vascular centerline is extracted from the three-dimensional vascular model, and the centerline nodes matching each target terminal point in the target terminal point set are determined. Path planning is performed on the target terminal point set based on the improved A-Star algorithm to obtain the optimal centerline path, and three-dimensional navigation is performed according to the optimal centerline path. Thus, by traversing all possible trajectories, all postures of the interventional surgical instrument in the delivery process are simulated, thereby solving the limitation of the simulation results caused by the single posture simulation process in the prior art. When traversing all possible trajectories, multiple trajectory screenings are performed to reduce the amount of calculation in the simulation process, and the improved A-Star algorithm is used to achieve the effect of quickly and accurately obtaining the optimal centerline path.

[0089] Optionally, in another embodiment disclosed in the present invention, referring to Figure 2 , Figure 2 The second flow chart of the three-dimensional navigation method of the vascular interventional surgery robot provided by the present invention is as follows: Figure 2 As shown: Based on the curvature value of each node on each delivery trajectory, each delivery trajectory is screened to obtain a target number of feature nodes, specifically including:

[0090] Step 2001, screening out a number of target nodes on each delivery trajectory whose curvature value is greater than or equal to a preset curvature value;

[0091] Specifically, the preset curvature value is set based on the characteristics of the interventional device. Since the support of the blood vessel wall is limited, the interventional device will cause damage to the blood vessel wall if it exceeds a certain force. The curvature value that the interventional device can be bent under a specific blood vessel support force is limited. Therefore, in the present invention, a minimum protection curvature value is set according to the characteristics of the interventional device, and the minimum protection curvature value is set as the preset curvature value.

[0092] In the present invention, after the curvature value of each node is obtained based on the above curvature value calculation formula, when the curvature value of the node is less than the preset curvature value, the node is discarded, and only several target nodes with curvature values ​​greater than or equal to the preset curvature value are retained. In this way, the delivery trajectory can be effectively constrained without affecting the efficiency of the path simulation.

[0093] Step 2002, determining a target number of characteristic nodes on each delivery trajectory according to the delivery displacement length of the surgical robot;

[0094] Specifically, the delivery displacement length of the surgical robot refers to the delivery distance of the interventional instrument of the surgical robot from the puncture starting point to the puncture end point.

[0095] During the delivery process, when the interventional device is not squeezed by the blood vessel wall, it will be affected by the elastic potential energy and will maintain a large range of straight line movement. Therefore, the present invention uses points with larger curvature as feature nodes of the trajectory to segment the delivery trajectory. In the present invention, the target number of feature nodes used for segmentation is N. seg The calculation formula is as follows:

[0096]

[0097] Where L refers to the delivery displacement length of the surgical robot, C L is a constant.

[0098] Step 2003, sorting a number of target nodes on each delivery trajectory from large to small based on the curvature value;

[0099] In this step, after retaining several target nodes whose curvature values ​​are greater than or equal to the preset curvature value, several target nodes on each delivery trajectory are numbered from large to small according to the curvature value, so that points with larger rates can be selected later as feature nodes of the trajectory.

[0100] Step 2004: The target nodes that are ranked in the front on each delivery track are used as feature nodes on each delivery track, and a target number of feature nodes are obtained.

[0101] Specifically, the numerical value between the target ranking and the target quantity is equal. For example, when the target quantity is calculated to be 10, the target nodes ranked in the top 10 after sorting are used as feature nodes on each delivery track.

[0102] The three-dimensional navigation method of a vascular interventional surgical robot proposed in the present invention screens out a number of target nodes on each delivery trajectory whose curvature values ​​are greater than or equal to a preset curvature value; then determines the target number of characteristic nodes on each delivery trajectory according to the delivery displacement length of the surgical robot; sorts the several target nodes on each delivery trajectory from large to small based on the curvature value; and uses the target nodes on each delivery trajectory that are ranked at the top as characteristic nodes on each delivery trajectory, thereby obtaining a target number of characteristic nodes, thereby effectively constraining the delivery trajectory without affecting the efficiency of path simulation.

[0103] Optionally, in another embodiment disclosed in the present invention, referring to Figure 3 , Figure 3 The third flow chart of the three-dimensional navigation method of the vascular interventional surgery robot provided by the present invention is as follows: Figure 3 As shown: the method of selecting a plurality of terminal points from a plurality of delivery trajectories based on the target number of characteristic nodes on each delivery trajectory, and selecting a target terminal point set from the plurality of terminal points, specifically includes:

[0104] Step 3001, dividing each delivery trajectory into a plurality of elastic rod trajectory segments based on a target number of characteristic nodes on each delivery trajectory, and calculating a total elastic potential energy value of each delivery trajectory based on the plurality of elastic rod trajectory segments;

[0105] Specifically, the total elastic potential energy value is the sum of the bending potential energy and the torsional potential energy of each elastic rod trajectory. In the present invention, a pseudo-rigid-body model is used to approximate the interventional device composed of elastic rods using a rigid body configuration with an equivalent force-deformation relationship, thereby calculating the bending potential energy of each elastic rod trajectory based on the pseudo-rigid-body model, and calculating the torsional potential energy of each elastic rod trajectory based on Kirchhoff's elastic rod theory.

[0106] Specifically, the total elastic potential energy value of each delivery trajectory is calculated based on a plurality of elastic rod trajectories, specifically including:

[0107] Step a, calculating the bending potential energy of each segment of the elastic rod trajectory based on a pseudo-rigid body model;

[0108] In practical applications, a pseudo rigid body model is first constructed, such as Figure 4 As shown, Figure 4 A schematic diagram of a pseudo-rigid body model of an interventional surgical instrument provided in an embodiment of the present invention, such as Figure 4As shown in the figure, the pseudo-rigid body model consists of three rigid rods, among which there is a translation pair and a rotation pair between the three rigid rods, and springs and torsion springs are added between them. Figure 4 It can be seen that the terminal change rate of the pseudo-rigid body model is Q = (Q x ,Q y ) T =(a / l,b / l) T ;

[0109]

[0110] Among them, y1 is the characteristic radius coefficient of each rigid rod, satisfying y0+y1=1, Δy0 is the variable characteristic radius coefficient, K θ is the spring constant, K L is the torsion spring stiffness coefficient.

[0111] In the present invention, the terminal change rate of the pseudo-rigid body model is first calculated, and then the following formula is calculated according to the terminal change rate:

[0112]

[0113] Wherein, θ is the end inclination angle of the pseudo-rigid body model. In the present invention, the end inclination angle of the pseudo-rigid body model is approximately equal to the end inclination angle of the compliant rod. Therefore, after obtaining the end inclination angle θ of the pseudo-rigid body model, based on the spring stiffness coefficient K θ and torsion spring stiffness coefficient K L The bending potential energy of each segment of the elastic rod trajectory can be calculated, where the spring stiffness coefficient K θ and torsion spring stiffness coefficient K L The bending potential energy E of each elastic rod trajectory can be calculated I The calculation formula is as follows:

[0114]

[0115] Where F is the resultant force at the end of the pseudo-rigid body model; l is the length of the compliant rod when it is not deformed; is the angle between the direction of action of the resultant force and the horizontal direction.

[0116] Step b, calculating the torsional potential energy of each section of the elastic rod trajectory based on the elastic rod theory;

[0117] Specifically, the calculation formula of torsional potential energy is as follows:

[0118]

[0119] Among them, β∈R 2＊2 , is the torsional stiffness constant of the interventional device; r is the radius value of the interventional device, m is the current torsion angle of the interventional device, and G is a constant.

[0120] Step c: calculating the total elastic potential energy value of each delivery trajectory according to the bending potential energy and the torsional potential energy of each segment of the elastic rod trajectory.

[0121] After the bending potential energy and the torsional potential energy of each section of the elastic rod trajectory are calculated, the sum of the bending potential energy and the torsional potential energy of each section of the elastic rod trajectory is calculated, thereby obtaining the total elastic potential energy value.

[0122] Step 3002, selecting at least one valid delivery trajectory from a plurality of delivery trajectories based on the total elastic potential energy value, and selecting a plurality of terminal points from the at least one valid delivery trajectory;

[0123] It is easy to understand that when the total elastic potential energy value of the delivery trajectory is too large, it indicates that the delivery trajectory will be strongly squeezed by the inner wall of the blood vessel. In other words, the collision type between the delivery trajectory and the inner wall of the blood vessel is a strong collision, which will cause damage to the blood vessel wall.

[0124] In the present invention, when the total elastic potential energy value of the delivery trajectory is greater than the preset total elastic potential energy threshold, the delivery trajectory is determined to be an invalid path trajectory, and when the total elastic potential energy value of the delivery trajectory is less than or equal to the preset total elastic potential energy threshold, the delivery trajectory is determined to be a valid path trajectory. Then, the terminal point within the preset range of the puncture end point is screened out from the retained multiple valid path trajectories.

[0125] Step 3003: clustering the plurality of terminal points into at least one type of terminal point set based on a clustering algorithm;

[0126] Specifically, the DBSCAN algorithm can be used in the present invention to divide an area with a sufficiently high density of terminal points into various types of terminal point sets. Specifically, when the DBSCAN algorithm is used to cluster a number of terminal points into at least one type of terminal point set, the two parameters of the scanning radius (eps) and the minimum number of included points (minPts) in the DBSCAN algorithm can be modified to cluster the several terminal points into different terminal point sets.

[0127] Step 3004: Filter out a target terminal point set from the at least one type of terminal point set according to the weighted average position of each type of terminal point set.

[0128] Specifically, the terminal point set is weighted averaged according to the bending potential energy of the delivery trajectory, and the weighted average position of the i-th type of terminal point set is recorded as P avg (i), where the weighted average position of the i-th terminal point set is calculated as follows:

[0129]

[0130] Where Cu(i) is the number of terminal points in the i-th terminal point set, Pi (n) is the three-dimensional coordinate value of the nth point in the i-th terminal point set, E i (n) is the bending potential energy of the nth point in the i-th cluster.

[0131] In the present invention, after calculating the weighted average position of each type of terminal point set, based on the position of the puncture end point, the terminal point set closest to the puncture end point position is selected as the target terminal point set.

[0132] The three-dimensional navigation method of a vascular interventional surgical robot proposed in the present invention first divides each delivery trajectory into a number of elastic rod trajectory segments based on the target number of characteristic nodes on each delivery trajectory, and calculates the total elastic potential energy value of each delivery trajectory based on the several elastic rod trajectory segments; then, based on the total elastic potential energy value, at least one valid path trajectory is screened out from the several delivery trajectories, and a number of terminal points are screened out from at least one valid delivery trajectory; finally, based on a clustering algorithm, the several terminal points are clustered into at least one type of terminal point set; and a target terminal point set is screened out from at least one type of terminal point set according to the weighted average position of each type of terminal point set, thereby successively eliminating invalid delivery trajectories from all delivery trajectories through a pseudo-rigid body model and Kirchhoff's elastic rod theory, and finally, based on a clustering algorithm, the target terminal point set with the highest probability is screened out, thereby ensuring that the optimal centerline path is accurately obtained.

[0133] Optionally, in another embodiment disclosed in the present invention, referring to Figure 5 , Figure 5 The fourth flow chart of the three-dimensional navigation method of the vascular interventional surgery robot provided by the present invention is as follows: Figure 5 As shown: the determining of the center line nodes matching each target terminal point in the target terminal point set specifically includes:

[0134] Step 4001, selecting a plurality of centerline nodes from the centerline of the blood vessel;

[0135] Specifically, a number of centerline nodes may be screened out from the blood vessel centerline at equal intervals, or a number of centerline nodes may be screened out based on multiple blood vessel branches of the blood vessel centerline, and the present invention is not limited to this.

[0136] Step 4002, selecting a set of centerline nodes located at the target position of each target terminal point from the plurality of centerline nodes;

[0137] Specifically, the target orientation refers to the orientation in the direction from the target terminal point to the puncture endpoint. In practical applications, the target terminal point can be first projected onto the blood vessel centerline. In the present invention, for ease of explanation, the point where the target terminal point is projected onto the blood vessel centerline is called the terminal projection point, and then the centerline node between the terminal projection point and the puncture endpoint is screened out from the blood vessel centerline.

[0138] Step 4003, respectively calculating the node distance between each centerline node in the centerline node set of each target terminal point and each target terminal point;

[0139] In practical applications, the node distances between each centerline node and each target terminal point are calculated based on the three-dimensional coordinates of each centerline node and each target terminal point in the three-dimensional blood vessel model.

[0140] Step 4004, sorting the centerline nodes in the centerline node set of each target terminal point from small to large based on the node distance, and taking the centerline nodes in a preset ranking as the centerline nodes matched by each target terminal point.

[0141] In practical applications, in order to ensure that nodes are not too concentrated on the center line in large-sized blood vessel planning, resulting in non-optimal path planning results, a node forward range k is set in the present invention to constrain the selection of centerline nodes.

[0142] Specifically, after obtaining the node distance between each centerline node and each target terminal point, the centerline node at the preset rank k is selected as the centerline node matching each target terminal point, wherein the value of k can be flexibly determined based on actual conditions. For example, when the centerline nodes on the centerline of the blood vessel are too dense, the value of the preset rank k is k1; when the centerline nodes on the centerline of the blood vessel are too dense, the value of the preset rank k is k2, wherein k1 is smaller than k2.

[0143] The three-dimensional navigation method of a vascular interventional surgical robot proposed in the present invention first selects a number of centerline nodes from the centerline of the blood vessel; then selects a centerline node set located at the target orientation of each target terminal point from the number of centerline nodes; calculates the node distance between each centerline node in the centerline node set of each target terminal point and each target terminal point; finally, based on the node distance, sorts each centerline node in the centerline node set of each target terminal point from small to large, and uses the centerline nodes in a preset ranking as the centerline nodes matched with each target terminal point, thereby ensuring that in the planning of large-size blood vessels, the nodes will not be too concentrated on the centerline, resulting in a non-optimal path planning result.

[0144] The following is a description of the three-dimensional navigation device for a vascular interventional surgery robot provided by the present invention. The three-dimensional navigation device for a vascular interventional surgery robot described below and the three-dimensional navigation method for a vascular interventional surgery robot described above can be referenced to each other.

[0145] refer to Figure 6 , Figure 6 FIG. 1 is a schematic diagram of the structure of the three-dimensional navigation device of the vascular intervention surgery robot provided by the present invention. Figure 6 As shown, the three-dimensional navigation device of the vascular interventional surgery robot includes: a traversal unit 610, which is used to determine the puncture starting point, the puncture end point and the target node set in the three-dimensional vascular model, and traverse the target node set to obtain a plurality of delivery trajectories, wherein the delivery trajectory is a trajectory starting from the puncture starting point, passing through the target node and finally reaching the puncture end point; a first screening unit 620, which is used to perform node screening on each delivery trajectory based on the curvature value of each node on each delivery trajectory, and obtain a target number of feature nodes; a second screening unit 630, which is used to screen a plurality of terminal points from the plurality of delivery trajectories based on the target number of feature nodes on each delivery trajectory, and screen a target terminal point set from the plurality of terminal points; an extraction unit 640, which is used to extract a plurality of terminal points from the plurality of terminal points in the three-dimensional vascular model. The centerline of the blood vessel is extracted from the model, and the centerline nodes matching each target terminal point in the target terminal point set are determined, and the centerline nodes are nodes located on the centerline of the blood vessel; the planning unit 650 is used to perform path planning on the target terminal point set based on the improved A-Star algorithm to obtain the optimal centerline path, and perform three-dimensional navigation according to the optimal centerline path; the total cost calculation formula of the improved A-Star algorithm is as follows: f(n)=g(n)+h′(n); wherein f(n) represents the total cost of the target terminal point n, g(n) represents the cost of the target terminal point n from the puncture starting point, and h′(n) represents the improved Euclidean distance of the target terminal point n from the puncture end point; the cost calculation formula of h′(n) is as follows: Among them, the Indicates the distance between the target terminal point n and the center line node se i The Euclidean distance, Indicates the centerline node se i The centerline distance from the puncture endpoint.

[0146] Furthermore, the planning unit 650 is also used to calculate the total cost of each target terminal point in the target terminal point set based on the improved A-Star algorithm; determine the target terminal point with the minimum total cost, and plan the optimal centerline path based on the target terminal point with the minimum total cost. Further, the planning unit 650 is also used to determine the radius of the optimized spherical model according to the delivery displacement length of the surgical robot; construct an optimized spherical model at the target terminal point set based on the radius of the optimized spherical model; and optimize the target terminal point set based on the optimized spherical model.

[0147] The three-dimensional navigation device of the vascular interventional surgery robot proposed in the present invention determines the puncture starting point, the puncture end point and the target node set in the three-dimensional vascular model, and traverses the target node set to obtain a plurality of delivery trajectories, then performs node screening on each delivery trajectory based on the curvature value of each node on each delivery trajectory to obtain a target number of feature nodes, then screens a plurality of terminal points from the plurality of delivery trajectories based on the target number of feature nodes on each delivery trajectory, and screens a target terminal point set from the plurality of terminal points, finally extracts the vascular centerline in the three-dimensional vascular model, determines the centerline node matching each target terminal point in the target terminal point set, performs path planning on the target terminal point set based on the improved A-Star algorithm, obtains the optimal centerline path, and performs three-dimensional navigation according to the optimal centerline path, thereby traversing all possible trajectories to simulate all postures of the interventional surgical instrument in the delivery process, thereby solving the limitation of the simulation result caused by the single posture simulation process in the prior art, and performing multiple trajectory screening when traversing all possible trajectories, reducing the amount of calculation in the simulation process, and using the improved A-Star algorithm to achieve the effect of quickly and accurately obtaining the optimal centerline path.

[0148] According to the three-dimensional navigation device of the vascular interventional surgical robot proposed in the present invention, the first screening unit 620 is also used to screen out a number of target nodes on each delivery trajectory whose curvature value is greater than or equal to a preset curvature value; determine the target number of characteristic nodes on each delivery trajectory according to the delivery displacement length of the surgical robot; and sort the several target nodes on each delivery trajectory from large to small based on the curvature value.

[0149] The three-dimensional navigation device of the vascular interventional surgical robot proposed in the present invention selects a number of target nodes on each delivery trajectory whose curvature value is greater than or equal to a preset curvature value; then determines the target number of characteristic nodes on each delivery trajectory according to the delivery displacement length of the surgical robot; sorts the several target nodes on each delivery trajectory from large to small based on the curvature value; and uses the target nodes on each delivery trajectory that are ranked at the top as characteristic nodes on each delivery trajectory, thereby obtaining a target number of characteristic nodes, thereby effectively constraining the delivery trajectory without affecting the path simulation efficiency.

[0150] According to the three-dimensional navigation device of the vascular interventional surgery robot proposed in the present invention, the second screening unit 630 is also used to divide each delivery trajectory into several segments of elastic rod trajectories based on the target number of characteristic nodes on each delivery trajectory, and calculate the total elastic potential energy value of each delivery trajectory based on the several segments of elastic rod trajectories; based on the total elastic potential energy value, at least one effective delivery trajectory is screened out from the several delivery trajectories, and several terminal points are screened out from at least one effective delivery trajectory; based on the clustering algorithm, the several terminal points are clustered into at least one type of terminal point set; according to the weighted average position of each type of terminal point set, the target terminal point set is screened out from the at least one type of terminal point set. Further, the second screening unit 630 is also used to calculate the bending potential energy of each segment of the elastic rod trajectory based on the pseudo-rigid body model; calculate the torsional potential energy of each segment of the elastic rod trajectory based on the elastic rod theory; and calculate the total elastic potential energy value of each delivery trajectory based on the bending potential energy of each segment of the elastic rod trajectory and the torsional potential energy.

[0151] The three-dimensional navigation device of the vascular interventional surgical robot proposed in the present invention first divides each delivery trajectory into a plurality of elastic rod trajectory segments based on the target number of characteristic nodes on each delivery trajectory, and calculates the total elastic potential energy value of each delivery trajectory based on the plurality of elastic rod trajectory segments; then selects at least one valid path trajectory from the plurality of delivery trajectories based on the total elastic potential energy value, and selects a plurality of terminal points from at least one valid delivery trajectory; finally, clusters the plurality of terminal points into at least one type of terminal point set based on a clustering algorithm; and selects a target terminal point set from at least one type of terminal point set according to the weighted average position of each type of terminal point set, thereby successively eliminating invalid delivery trajectories from all delivery trajectories through a pseudo-rigid body model and Kirchhoff's elastic rod theory, and finally selects the target terminal point set with the highest probability based on a clustering algorithm, thereby ensuring that the optimal centerline path is accurately obtained.

[0152] According to the three-dimensional navigation device of the vascular interventional surgery robot proposed in the present invention, the extraction unit 640 is also used to select a number of centerline nodes from the centerline of the blood vessel; select a centerline node set located at the target orientation of each target terminal point from the number of centerline nodes; calculate the node distance between each centerline node in the centerline node set of each target terminal point and each target terminal point; sort the centerline nodes in the centerline node set of each target terminal point from small to large based on the node distance, and use the centerline nodes located in preset ranks as the centerline nodes matched by each target terminal point.

[0153] The three-dimensional navigation device of the vascular interventional surgery robot proposed in the present invention first selects a plurality of centerline nodes from the centerline of the blood vessel; then selects a centerline node set located at the target orientation of each target terminal point from the plurality of centerline nodes; calculates the node distance between each centerline node in the centerline node set of each target terminal point and each target terminal point; finally, based on the node distance, sorts each centerline node in the centerline node set of each target terminal point from small to large, and uses the centerline nodes in a preset ranking as the centerline nodes matched with each target terminal point, thereby ensuring that in the planning of large-size blood vessels, the nodes will not be too concentrated on the centerline, resulting in a non-optimal path planning result.

[0154] Figure 7 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 7As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730 and a communication bus 740, wherein the processor 710, the communication interface 720 and the memory 730 communicate with each other through the communication bus 740. The processor 710 may call the logic instructions in the memory 730 to execute a three-dimensional navigation method for a vascular interventional surgery robot, the method comprising: determining a puncture starting point, a puncture end point and a target node set in a three-dimensional vascular model, and traversing the target node set to obtain a plurality of delivery trajectories, wherein the delivery trajectory is a trajectory starting from the puncture starting point, passing through the target node and finally reaching the puncture end point; performing node screening on each delivery trajectory based on the curvature value of each node on each delivery trajectory to obtain a target number of feature nodes; screening a plurality of terminal points from a plurality of delivery trajectories based on the target number of feature nodes on each delivery trajectory, and screening a target terminal point set from the plurality of terminal points; screening a plurality of terminal points in the three-dimensional vascular model ... The centerline of the blood vessel is taken out, and the centerline nodes matching each target terminal point in the target terminal point set are determined, and the centerline nodes are nodes located on the centerline of the blood vessel; the target terminal point set is path planned based on the improved A-Star algorithm to obtain the optimal centerline path, and three-dimensional navigation is performed according to the optimal centerline path; the total cost calculation formula of the improved A-Star algorithm is as follows: f(n)=g(n)+h′(n); wherein f(n) represents the total cost of the target terminal point n, g(n) represents the cost of the target terminal point n from the puncture starting point, and h′(n) represents the improved Euclidean distance of the target terminal point n from the puncture end point; the cost calculation formula of h′(n) is as follows: Among them, the O sei (n) represents the distance between the target terminal point n and the center line node se i The Euclidean distance, T sei Indicates the centerline node se i The centerline distance from the puncture endpoint.

[0155] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0156] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the three-dimensional navigation method of the vascular interventional surgery robot provided by the above-mentioned methods, and the method includes: determining the puncture starting point, the puncture end point and the target node set in the three-dimensional vascular model, and traversing the target node set to obtain a plurality of delivery trajectories, wherein the delivery trajectory is a trajectory starting from the puncture starting point, passing through the target node and finally reaching the puncture end point; performing node screening on each delivery trajectory based on the curvature value of each node on each delivery trajectory to obtain a target number of feature nodes; screening a plurality of terminal points from a plurality of delivery trajectories based on the target number of feature nodes on each delivery trajectory. , and select a target terminal point set from the plurality of terminal points; extract the vascular centerline from the three-dimensional vascular model, determine the centerline node matching each target terminal point in the target terminal point set, and the centerline node is a node located on the vascular centerline; perform path planning on the target terminal point set based on the improved A-Star algorithm to obtain an optimal centerline path, and perform three-dimensional navigation according to the optimal centerline path; the total cost calculation formula of the improved A-Star algorithm is as follows: f(n)=g(n)+h′(n); wherein f(n) represents the total cost of the target terminal point n, g(n) represents the cost of the target terminal point n from the puncture starting point, and h′(n) represents the improved Euclidean distance of the target terminal point n from the puncture end point; the cost calculation formula of h′(n) is as follows: Among them, the Indicates the distance between the target terminal point n and the center line node se i The Euclidean distance, Indicates the centerline node se iThe centerline distance from the puncture endpoint.

[0157] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the three-dimensional navigation method of a vascular interventional surgical robot provided by the above-mentioned methods, the method comprising: determining a puncture starting point, a puncture end point and a target node set in a three-dimensional vascular model, and traversing the target node set to obtain a plurality of delivery trajectories, wherein the delivery trajectory is a trajectory starting from the puncture starting point, passing through the target node and finally arriving at the puncture end point; performing node screening on each delivery trajectory based on the curvature value of each node on each delivery trajectory to obtain a target number of feature nodes; screening a plurality of terminal points from the plurality of delivery trajectories based on the target number of feature nodes on each delivery trajectory, and screening out a plurality of terminal points from the plurality of terminal points. A target terminal point set; extracting a blood vessel centerline from the three-dimensional blood vessel model, determining a centerline node matching each target terminal point in the target terminal point set, wherein the centerline node is a node located on the centerline of the blood vessel; performing path planning on the target terminal point set based on the improved A-Star algorithm to obtain an optimal centerline path, and performing three-dimensional navigation according to the optimal centerline path; the total cost calculation formula of the improved A-Star algorithm is as follows: f(n)=g(n)+h′(n); wherein f(n) represents the total cost of the target terminal point n, g(n) represents the cost of the target terminal point n from the puncture starting point, and h′(n) represents the improved Euclidean distance of the target terminal point n from the puncture end point; the cost calculation formula of h′(n) is as follows: Among them, the Indicates the distance between the target terminal point n and the center line node se i The Euclidean distance, Indicates the centerline node se i The centerline distance from the puncture endpoint.

[0158] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0159] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0160] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A three-dimensional navigation method for a vascular interventional surgery robot, characterized in that: include: Determine the puncture starting point, puncture end point and target node set in the three-dimensional blood vessel model, and traverse the target node set to obtain a plurality of delivery trajectories, wherein the delivery trajectory is a trajectory starting from the puncture starting point, passing through the target node and finally reaching the puncture end point; Based on the curvature value of each node on each delivery trajectory, each delivery trajectory is screened to obtain a target number of feature nodes; Filtering a plurality of terminal points from a plurality of delivery trajectories based on a target number of characteristic nodes on each delivery trajectory, and filtering a target terminal point set from the plurality of terminal points; Extracting a blood vessel centerline in the three-dimensional blood vessel model, and determining a centerline node matching each target terminal point in the target terminal point set, wherein the centerline node is a node located on the blood vessel centerline; Performing path planning on the target terminal point set based on the improved A-Star algorithm to obtain an optimal centerline path, and performing three-dimensional navigation according to the optimal centerline path; The total cost calculation formula of the improved A-Star algorithm is as follows: f(n)=g(n)+h′(n) Wherein, f(n) represents the total cost of the target terminal point n, g(n) represents the cost of the target terminal point n from the puncture starting point, and h′(n) represents the improved Euclidean distance between the target terminal point n and the puncture end point; The cost calculation formula of h′(n) is as follows: Among them, the Indicates the distance between the target terminal point n and the center line node se i The Euclidean distance, Indicates the centerline node se i The centerline distance from the puncture endpoint.

2. The three-dimensional navigation method of a vascular interventional surgery robot according to claim 1, characterized in that: The determining of the centerline node matching each target terminal point in the target terminal point set specifically includes: Selecting a plurality of centerline nodes from the centerline of the blood vessel; Selecting a set of centerline nodes located at the target position of each target terminal point from the plurality of centerline nodes respectively; Calculate the node distances between each centerline node in the centerline node set of each target terminal point and each target terminal point respectively; The centerline nodes in the centerline node set of each target terminal point are sorted from small to large based on the node distance, and the centerline nodes in the preset ranking are used as the centerline nodes matched by each target terminal point.

3. The three-dimensional navigation method of a vascular interventional surgery robot according to claim 1, characterized in that: The path planning of the target terminal point set based on the improved A-Star algorithm to obtain the optimal centerline path specifically includes: Calculate the total cost of each target terminal point in the target terminal point set based on the improved A-Star algorithm; A target terminal point with the minimum total cost is determined, and an optimal centerline path is planned based on the target terminal point with the minimum total cost.

4. The three-dimensional navigation method of a vascular interventional surgery robot according to claim 1, characterized in that: The node screening of each delivery trajectory is performed based on the curvature value of each node on each delivery trajectory to obtain a target number of feature nodes, specifically including: Filter out a number of target nodes on each delivery trajectory whose curvature value is greater than or equal to a preset curvature value; Determining a target number of characteristic nodes on each delivery trajectory according to the delivery displacement length of the surgical robot; Sort several target nodes on each delivery trajectory from large to small based on the curvature value; The target nodes ranked in the front on each delivery trajectory are respectively used as feature nodes on each delivery trajectory, and the target number of feature nodes are obtained.

5. The three-dimensional navigation method of a vascular interventional surgery robot according to claim 1, characterized in that: The step of selecting a plurality of terminal points from a plurality of delivery trajectories based on a target number of characteristic nodes on each delivery trajectory, and selecting a target terminal point set from the plurality of terminal points, specifically includes: Based on the target number of characteristic nodes on each delivery trajectory, each delivery trajectory is divided into a plurality of elastic rod trajectories, and a total elastic potential energy value of each delivery trajectory is calculated based on the plurality of elastic rod trajectories; Selecting at least one effective delivery trajectory from the plurality of delivery trajectories based on the total elastic potential energy value, and selecting a plurality of terminal points from the at least one effective delivery trajectory; Clustering the plurality of terminal points into at least one type of terminal point set based on a clustering algorithm; A target terminal point set is screened out from the at least one type of terminal point set according to the weighted average position of each type of terminal point set.

6. The three-dimensional navigation method of a vascular interventional surgery robot according to claim 5, characterized in that: The total elastic potential energy value of each delivery trajectory is calculated based on a plurality of elastic rod trajectories, specifically including: The bending potential energy of each segment of the elastic rod trajectory is calculated based on the pseudo-rigid body model; The torsional potential energy of each elastic rod trajectory is calculated based on Kirchhoff's elastic rod theory; The total elastic potential energy value of each delivery trajectory is calculated according to the bending potential energy and the torsional potential energy of each segment of the elastic rod trajectory.

7. The three-dimensional navigation method of a vascular interventional surgery robot according to any one of claims 1 to 6, characterized in that: Before performing path planning on the target terminal point set based on the improved A-Star algorithm to obtain the optimal centerline path, the method further includes: Determining the radius of the optimized spherical model according to the delivery displacement length of the surgical robot; constructing an optimized spherical model at the target terminal point set based on the radius of the optimized spherical model; The target terminal point set is optimized based on the optimized spherical model.

8. A three-dimensional navigation device for a vascular interventional surgery robot, characterized in that: include: A traversal unit, used to determine the puncture starting point, the puncture end point and the target node set in the three-dimensional blood vessel model, and traverse the target node set to obtain a plurality of delivery trajectories, wherein the delivery trajectory is a trajectory starting from the puncture starting point, passing through the target node and finally reaching the puncture end point; A first screening unit is used to screen nodes on each delivery trajectory based on the curvature value of each node on each delivery trajectory to obtain a target number of feature nodes; A second screening unit is used to screen out a plurality of terminal points from the plurality of delivery trajectories based on a target number of characteristic nodes on each delivery trajectory, and screen out a target terminal point set from the plurality of terminal points; An extraction unit, configured to extract a blood vessel centerline from the three-dimensional blood vessel model, and determine a centerline node matching each target terminal point in the target terminal point set, wherein the centerline node is a node located on the blood vessel centerline; A planning unit, configured to perform path planning for the target terminal point set based on an improved A-Star algorithm, obtain an optimal centerline path, and perform three-dimensional navigation according to the optimal centerline path; The total cost calculation formula of the improved A-Star algorithm is as follows: f(n)=g(n)+h′(n) Wherein, f(n) represents the total cost of the target terminal point n, g(n) represents the cost of the target terminal point n from the puncture starting point, and h′(n) represents the improved Euclidean distance between the target terminal point n and the puncture end point; The cost calculation formula of h′(n) is as follows: Among them, the Indicates the distance between the target terminal point n and the center line node se i The Euclidean distance, Indicates the centerline node se i The centerline distance from the puncture endpoint.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the three-dimensional navigation method of the vascular interventional surgery robot is implemented as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the three-dimensional navigation method of a vascular interventional surgery robot is implemented as described in any one of claims 1 to 7.

Images