A robot-assisted fracture reduction path planning method and system

Abstract

The present application discloses a robot-assisted fracture reduction path planning method and system, which are applied to the field of fracture reduction technology. The method includes: acquiring CT data of a patient; segmenting, reconstructing and planning target positions of the CT data to obtain initial position data and target position data corresponding to the broken bone; calculating the position difference and posture difference between the initial position data and the target position data; using the initial position data and the target position data to form a six-dimensional vector as a sampling point, performing fracture reduction path planning to generate a final path; and optimizing the final path through a path post-processing algorithm.

Description

Technical Field

[0001] The present invention relates to the technical field of fracture reduction, and in particular to a robot-assisted fracture reduction path planning method and system. Background Art

[0002] With the continuous development of technology, robotic-assisted minimally invasive orthopedic surgery has become an increasingly popular surgical method. Unlike traditional surgical methods, robotic surgery can intuitively display individual anatomical characteristics through three-dimensional image analysis and preoperative planning, helping doctors perform accurate operations. It significantly reduces the number of operating steps in key surgical links, alleviates soft tissue interference, shortens the operation time, and minimizes the risk of surgical trauma and related complications. Compared with its widespread application in spinal surgery and joint replacement surgery, robotic technology has not yet been widely used in pelvic fracture reduction surgery. Computer-assisted preoperative planning is crucial for pelvic fracture reduction surgery. It can design the optimal surgical path and surgical plan, improving the safety and success rate of the operation. From the perspective of robotic autonomous surgery, preoperative planning can quantify the fracture reduction path to safely avoid collisions between bone fragments.

[0003] Pan et al. [1] used an enhanced A* algorithm for pelvic fracture reduction path planning and proposed a direction planning strategy and linear sampling search to optimize the reduction path. In this method, the spatial position of the bone fragments was traversed and searched using a three-dimensional A* algorithm, and the spatial posture of the bone fragments was planned using a linear sampling method. The scheme also established an in vitro pelvic experimental platform to verify the feasibility of the algorithm. In addition, the team also combined OpenSim software with this method to explore reducing muscle resistance during the reduction process [2].

[0004] [1]Pan M, Chen Y, Li Z, et al. Autonomous path planning for robot-assisted pelvic fracture closed reduction with collision avoidance[J]. The International Journal of Medical Robotics and Computer Assisted Surgery, 2023, 19(2):e2483.

[0005] [2]Chen Y, Pan MZ, Li Z, et al. Autonomous path planning with muscleforce optimization for robot assisted pelvic fracture closed reduction[J]. IEEE Transactions on Medical Robotics and Bionics, 2023.

[0006] Among existing pelvic fracture reduction path planning methods, the A* algorithm suffers from low planning efficiency due to the need to traverse three-dimensional spatial points. Furthermore, a linear planning strategy is required for the spatial posture of the broken bones, which makes it impossible to successfully plan paths for complex fracture cases in the confined space of the pelvis.

[0007] To overcome these shortcomings, this application proposes a robot-assisted fracture reduction path planning method and system. This system simultaneously samples and searches the three-dimensional spatial position and three-dimensional spatial posture of the bone fragments, combines a surface point cloud-based collision detection algorithm with a path shorting algorithm, and provides a quantitative path for robot-assisted fracture reduction. Summary of the Invention

[0008] The purpose of this application is to provide a robot-assisted fracture reduction path planning method and system to solve the above-mentioned problems.

[0009] To achieve the above objectives, this application provides the following technical solutions:

[0010] The present application provides a robot-assisted fracture reduction path planning method, comprising:

[0011] Obtain CT data of the patient;

[0012] Segmenting and reconstructing the CT data and planning the target position to obtain initial position data and target position data corresponding to the broken bone;

[0013] Calculating a position difference and a posture difference between the initial position data and the target position data;

[0014] The initial position data and the target position data form a six-dimensional vector as a sampling point, perform fracture reduction path planning, and generate a final path;

[0015] The final path is optimized by a path post-processing algorithm.

[0016] Furthermore, the step of segmenting and reconstructing the CT data and planning the target position data to obtain the initial position data and target position data corresponding to the broken bone specifically includes the following steps:

[0017] Segment and reconstruct the CT data to form a configuration space containing the broken bones The configuration space is divided into two categories: healthy bones and bones that do not need to be moved Defined as obstacle area, other areas The initial position data of the broken bone is defined as χ start ;

[0018] Select the target position data x for the broken bone target , restore the physiological state of the pelvic structure.

[0019] Furthermore, the step of calculating the position difference and posture difference between the initial position data and the target position data specifically includes the following steps:

[0020] Let the Euler angle at the initial posture be θ s =(0,0,0), then the Euler angle of the target posture is θ t =(α t ,β t ,γ t );

[0021] The initial pose data and the target pose data are defined as vectors of length 6 respectively, then:

[0022]

[0023] If at time τ and in the Cartesian coordinate system, the center of mass of the broken bone moves to (x τ ,y τ ,z τ ), the Euler angles are rotated by (α τ ,β τ ,γ τ ), then the initial position of the bone point cloud F τ Expressed as:

[0024] F τ =r Z (γ τ )r Y (β τ )r X (α τ )·(F s -P s )+(x τ ,y τ ,z τ ).

[0025] Furthermore, the step of using the initial position data and the target position data to form a six-dimensional vector as sampling points to perform fracture reduction path planning and generate a final path specifically includes the following steps:

[0026] Use the initial position as the node of the tree and build a tree including the starting point;

[0027] By iterating the enhanced RRT* algorithm, a node χ is randomly sampled from the state space. rand , and find the node χ closest to the node in the tree nearest ;

[0028] The node χ is transformed into rand Expand to a node with a step length χ new ;

[0029] If we judge χ new and χ nearest If the line between them collides with an obstacle, the node is abandoned;

[0030] For the nodes that are not abandoned, select a list of all neighboring nodes on the tree whose distance is less than the preset threshold X near , and perform two optimization steps.

[0031] Furthermore, for the nodes that are not abandoned, a list X of all neighboring nodes whose distance is less than a preset threshold is selected on the tree. near , the two optimization steps specifically include the following steps:

[0032] In the first optimization step, we select χ new The parent node of the node is calculated from the starting point through each neighboring node χ near Arrival new Cost;

[0033] The neighboring node χ on the shortest cost path near Let χ new The parent node of new Add to the tree and become a node on the tree;

[0034] The second optimization step is to select the neighboring node χ near The parent node of new Arrival at neighboring node χ near the price;

[0035] If through χ new Arrival near The cost is less than the neighboring node χ near The original cost of χ near The parent node is set to χ new ;

[0036] When the maximum number of iterations is reached or the end condition is met, a reverse search starts from the target point to generate the final path.

[0037] Furthermore, the enhanced RRT* algorithm also includes:

[0038] The nodes are expanded by synchronously exploring asynchronous adjustments, and the position coordinates and Euler angles are adjusted separately with steps of Δ1 and Δ2 respectively.

[0039] The collision detection algorithm based on the surface point cloud of the object is used to determine the moving bone F temp With other bone blocks F obs Whether there is a collision between them is determined by detecting whether there is an intersection S between the point clouds:

[0040] S=F temp ∩F obs

[0041] Traverse F temp For each point in F, calculate the obs The nearest neighbor points in ;

[0042] If there are neighboring points within the preset threshold δ, it is determined that a collision occurs, otherwise it is determined that no collision occurs;

[0043] The time complexity of the collision detection algorithm is O(n 2 ), KD tree is used to model the point cloud of environmental obstacles, reducing the time complexity to O(n log n).

[0044] Furthermore, the step of optimizing the final path by using a path post-processing algorithm specifically includes the following steps:

[0045] Starting from the starting point, check whether the Lth point is directly connected to the starting point and there is no collision with obstacles on the path; if there is no collision, directly connect and delete the intermediate nodes; if there is a collision, check the L-1th point until a collision-free node is found; starting from this node, repeat the test until reaching the target point, and finally return the remaining nodes and edges;

[0046] The path processed by the path short-circuiting algorithm retains the n+1 key nodes V that the broken bones must pass through during the reduction process. i , after connecting end to end, generate n geometrically continuous lines corresponding to n intervals; smooth the line segments connected by key nodes to obtain a space curve that meets the parameter continuity conditions, where the parameter continuity includes geometric continuity, velocity continuity and acceleration continuity; specifically:

[0047] The cubic spline interpolation method is used to segmentally smooth the reset path. The segmented parametric curve in three-dimensional space is defined as:

[0048] Γ(τ)={Γ i (τ),τ∈[τ i ,τ i+1 ],i=0,…,n-1}

[0049] Where τ0=0, τ n =1; corresponding to the interpolation coordinates in each interval, the cubic spline interpolation definition is used:

[0050]

[0051] Among them, there are 12n unknown parameters in n intervals, and they are solved through 12n equations;

[0052] The specific constraints include: the function values at the beginning and end and internal nodes are known, generating 6n equations; the first-order derivatives at adjacent interval nodes are equal, generating 3(n-1) equations; the second-order derivatives at adjacent interval nodes are equal, generating 3(n-1) equations; the second-order derivatives at the beginning and end of the entire path are zero, generating 6 equations; the formula is expressed as:

[0053]

[0054] The position coordinates V of the path nodes are i (p) and posture coordinates V i (θ) can be interpolated to obtain a smooth motion trajectory.

[0055] The present application provides a robot-assisted fracture reduction path planning system, comprising:

[0056] Acquisition module: acquires the patient's CT data;

[0057] Data processing module: performs segmentation, reconstruction and target position planning on the CT data to obtain initial position data and target position data corresponding to the broken bone; calculates the position difference and posture difference between the initial position data and the target position data;

[0058] Fracture reduction path planning module: using the initial position data and the target position data to form a six-dimensional vector as sampling points, performing fracture reduction path planning and generating a final path;

[0059] Optimization module: optimizes the final path through a path post-processing algorithm.

[0060] The present application provides a device, which includes a processor and a memory coupled to the processor, wherein the memory stores program instructions for implementing a robot-assisted fracture reduction path planning method; the processor is used to execute the program instructions stored in the memory to implement a robot-assisted fracture reduction path planning.

[0061] The present application provides a storage medium storing program instructions executable by a processor, wherein the program instructions are used to execute a robot-assisted fracture reduction path planning method.

[0062] This application provides a robot-assisted fracture reduction path planning method and system, which has the following beneficial effects:

[0063] (1) By enhancing the RRT algorithm, the robot can simultaneously plan the three-dimensional spatial position and three-dimensional spatial posture of the bone fragment in six-dimensional space, greatly improving the efficiency of path planning;

[0064] (2) A collision detection method based on the point cloud of the object surface is adopted, which can take into account the actual surface shape of the bone fragments and obstacles to ensure safety during the reduction process; at the same time, the path short-connection method and the path post-processing method of cubic spline interpolation are used to further ensure the smoothness of the path and shorten the path length, thereby optimizing the reduction path. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 This is a flow chart of a robot-assisted fracture reduction path planning method according to Example 1 of the present application;

[0066] Figure 2 This is a flow chart of the second optimization step of Example 1 of the present application;

[0067] Figure 3 Schematic diagram of the flow of the path growth algorithm of Example 1 of the present application;

[0068] Figure 4 This is a flowchart of the collision detection algorithm of Example 1 of the present application;

[0069] Figure 5 This is a schematic diagram of the path short-circuiting of Example 1 of the present application;

[0070] Figure 6 This is a flow chart of the path post-processing algorithm of Example 1 of the present application;

[0071] Figure 7 Schematic diagram of path planning results under different types of cases in Example 1 of this application;

[0072] Figure 8 This is a schematic diagram of the quantification results of Example 1 of the present application;

[0073] Figure 9 This is a schematic diagram of the quantitative comparison results of different solutions in Example 1 of the present application;

[0074] Figure 10 This is a schematic structural diagram of a robot-assisted fracture reduction path planning system according to Example 2 of the present application;

[0075] Figure 11 This is a schematic diagram of the device structure of Example 3 of the present application;

[0076] Figure 12 This is a schematic diagram of the storage medium structure of Example 4 of the present application. DETAILED DESCRIPTION

[0077] It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application.

[0078] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0079] Example 1

[0080] See also Figure 1 , is a flow chart of a robot-assisted fracture reduction path planning method according to Example 1 of the present application; the steps include:

[0081] S1: Acquire CT data of the patient.

[0082] S2: Segment and reconstruct the CT data and perform target position planning to obtain initial position data and target position data corresponding to the broken bone.

[0083] In this embodiment, the CT data is segmented and reconstructed to form a configuration space containing the broken bones. The configuration space is divided into two categories: healthy bones and bones that do not need to be moved Defined as obstacle area, other areas The initial position data of the broken bone is defined as χ start .

[0084] Select the target position data x for the broken bone target , restore the physiological state of the pelvic structure and provide a complete pelvic effect after reconstruction.

[0085] S3: Calculate the position difference and posture difference between the initial position data and the target position data.

[0086] In this embodiment, in rigid body dynamics, the three-dimensional motion of a rigid body can be expressed as the combined motion of the translation of the rigid body's center of mass and the rotation of the rigid body around the center of mass. In the world coordinate system, let the Euler angle at the initial posture be θ s =(0,0,0), then the Euler angle of the target posture is θ t =(α t ,β t ,γ t ). Therefore, the point cloud pose can be expressed as a combination of the center of mass coordinates P and the Euler angle coordinates θ.

[0087] The initial pose data and the target pose data are defined as vectors of length 6 respectively, then:

[0088]

[0089] Although X is just a vector, it reflects the actual coordinates of the entire point cloud. During the reset process, the reset path can be represented by the trajectory of the center of mass, and each position also contains the Euler angle relative to the initial position at that moment. Therefore, if at time ω and in the Cartesian coordinate system, the center of mass of the broken bone moves to (x τ ,y τ ,z τ ), the Euler angles are rotated by (α τ ,β τ ,γ τ ), then the initial position of the bone point cloud F τ Expressed as:

[0090] F τ =R Z (γ τ )R Y (β τ )R X (α τ )·(F s -P s )+(x τ ,y τ ,z τ ).

[0091] S4: Using the initial position data and the target position data to form a six-dimensional vector as sampling points, perform fracture reduction path planning to generate a final path.

[0092] In this embodiment, the initial position is first used as the only node of the tree to build a tree containing the starting point. The enhanced RRT* algorithm is iterated to randomly sample a node x from the state space.rand , and find the node χ closest to the node in the tree nearest ; Then the node χ is transformed into rand Expand to a node with a step length χ new .

[0093] If we judge χ new and χ nearest If the line between them collides with an obstacle, the node is abandoned; for the nodes that are not abandoned, a list of all neighboring nodes with a distance less than a preset threshold is selected on the tree. near , and perform two optimization steps.

[0094] In the first optimization step, we select χ new The parent node of the node is calculated from the starting point through each neighboring node χ near Arrival new Cost; the neighboring node χ on the shortest cost path near Let χ new The parent node of new Add to the tree and become a node on the tree.

[0095] See also Figure 2 , which is a flow algorithm diagram of the second optimization step of Example 1 of the present application.

[0096] In the second optimization step, the neighboring nodes χ are selected near The parent node of new Arrival at neighboring node χ near The cost; if through χ new Arrival near The cost is less than the neighboring node χ near The original cost of χ near The parent node is set to χ new .

[0097] When the maximum number of iterations is reached or the end condition is met, a reverse search starts from the target point to generate the final path.

[0098] See also Figure 3 , which is a flow chart of the path growth algorithm of Example 1 of the present application.

[0099] The path growth algorithm is used to expand the node and return a new node χ new In the basic RRT* algorithm, the node expansion is from the current node to the random sampling point χ rand Grow a fixed step size to generate a new node. In order to consider the node expansion of both position and posture, this application proposes a synchronous exploration and asynchronous adjustment strategy, which adjusts the position coordinates and Euler angles separately with step sizes of Δ1 and Δ2 respectively.

[0100] See also Figure 4 , which is a flow chart of the collision detection algorithm of Example 1 of the present application.

[0101] The collision detection algorithm based on the surface point cloud of the object is used to determine the moving bone F temp With other bone blocks F obs Whether there is a collision between them is determined by detecting whether there is an intersection S between the point clouds:

[0102] S=F temp ∩F obs

[0103] Traverse F temp For each point in F, calculate the obs The nearest neighbor points in are expressed in Euclidean distance.

[0104] If there are neighboring points within the preset threshold δ, it is determined that a collision occurs; otherwise, it is determined that no collision occurs.

[0105] The time complexity of the collision detection algorithm is O(n 2 To improve computational efficiency, a KD tree is used to model the point cloud of environmental obstacles, reducing the computational complexity to O(n log n). Testing has shown that a single check in a model point cloud of approximately 10,000 points takes only 0.015 seconds.

[0106] The functions in the figure are explained as follows: the Linspace function is used to return a uniformly distributed six-dimensional vector between two postures; the FindNeighbor function represents KD tree modeling of the point cloud and finding the nearest neighbors of the input point; Norm is used to calculate the Euclidean distance between the coordinates of two three-dimensional points.

[0107] S5: Optimizing the final path through a path post-processing algorithm.

[0108] In this example, starting from the starting point, the Lth point is checked to see if it is directly connected to the starting point and does not collide with obstacles along the path. If there is no collision, the connection is made directly and the intermediate nodes are deleted. If there is a collision, the L-1th point is checked until a collision-free node is found. From this node, the test is repeated until the target point is reached, and the remaining nodes and edges are returned.

[0109] The path processed by the path short-circuiting algorithm retains the n+1 key nodes V that the broken bones must pass through during the reduction process. i, after connecting end to end, n geometrically continuous line segments are generated, corresponding to n intervals; the line segments connected by key nodes are smoothed to obtain a spatial curve that meets the parameter continuity conditions, where the parameter continuity includes geometric continuity, velocity continuity and acceleration continuity.

[0110] In addition, these key nodes have been subjected to anti-collision detection. To ensure the safety of the path, the smoothed curve should strictly pass through these key nodes. Therefore, the cubic spline interpolation method is used to smooth the reset path piecewise. The piecewise parametric curve in three-dimensional space is defined as:

[0111] Γ(τ)={Γ i (τ),τ∈[τ i ,τ i+1 ],i=0,…,n-1}

[0112] Where τ0=0, τ n =1; corresponding to the interpolation coordinates in each interval, the cubic spline interpolation definition is used:

[0113]

[0114] It can be seen from the above formula that there are 12 unknown parameters in each dimension in each interval. Therefore, there are 12n unknown parameters in n intervals, which are solved through 12n equations.

[0115] The specific constraints include: the function values at the beginning and end and internal nodes are known, generating 6n equations; the first-order derivatives at adjacent interval nodes are equal, generating 3(n-1) equations; the second-order derivatives at adjacent interval nodes are equal, generating 3(n-1) equations; the second-order derivatives at the beginning and end of the entire path are zero, generating 6 equations; the formula is expressed as:

[0116]

[0117] The position coordinates V of the path nodes are i (p) and posture coordinates V i (θ) can be interpolated to obtain a smooth motion trajectory.

[0118] In summary, Example 1 of the present application uses an enhanced six-dimensional RRT* algorithm to simultaneously plan the three-dimensional spatial position and three-dimensional spatial posture of the broken bones in six-dimensional space; and adopts a collision detection method based on the surface point cloud of the object, taking into account the actual surface shape of the broken bones and obstacles to ensure the safety of the patient and the robot during the reduction process; at the same time, for the planned path nodes, a path short-connection method and a path post-processing method of cubic spline interpolation are proposed to ensure the smoothness of the path and shorten the length of the path; it helps to improve the safety and success rate of the operation and reduce the risk of surgical trauma and related complications.

[0119] Example 2

[0120] See also Figure 10 , is a schematic diagram of the structure of a robot-assisted fracture reduction path planning system according to Example 2 of the present application; the specific contents include:

[0121] Acquisition module: acquires the patient's CT data;

[0122] Data processing module: performs segmentation, reconstruction and target position planning on the CT data to obtain initial position data and target position data corresponding to the broken bone; calculates the position difference and posture difference between the initial position data and the target position data;

[0123] Fracture reduction path planning module: using the initial position data and the target position data to form a six-dimensional vector as sampling points, performing fracture reduction path planning and generating a final path;

[0124] Optimization module: optimizes the final path through a path post-processing algorithm.

[0125] Example 3

[0126] See also Figure 11 , which is a schematic diagram of the device structure of Example 3 of the present application. The device 50 includes a processor 51 and a memory 52 coupled to the processor 51.

[0127] The memory 52 stores program instructions for implementing the above-mentioned robot-assisted fracture reduction path planning method.

[0128] The processor 51 is configured to execute program instructions stored in the memory 52 to implement robot-assisted fracture reduction path planning.

[0129] The processor 51 may also be referred to as a CPU (Central Processing Unit).

[0130] Processor 51 may be an integrated circuit chip with signal processing capabilities. Processor 51 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component. A general-purpose processor may be a microprocessor or any conventional processor.

[0131] Example 4

[0132] See also Figure 12, which is a structural diagram of the storage medium of Example 4 of the present application. The storage medium of the embodiment of the present application stores a program file 61 that can implement all the above methods, wherein the program file 61 can be stored in the above storage medium in the form of a software product, including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to execute all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or a computer, server, mobile phone, tablet and other devices.

[0133] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, apparatus, article, or method comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, apparatus, article, or method. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, apparatus, article, or method comprising the element.

[0134] The above description is only a preferred embodiment of the present application and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

[0135] Although the embodiments of the present application have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations may be made to these embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

[0136] Of course, the present invention may have many other implementations. Based on this implementation, other implementations obtained by ordinary technicians in this field without any creative work are all within the scope of protection of the present invention.

Claims

1. A robot-assisted fracture reduction path planning method, characterized in that: include: Obtain CT data of the patient; Performing segmentation and reconstruction and target posture planning on the CT data to obtain initial posture data and target posture data corresponding to the broken bone; Calculating a position difference and a posture difference between the initial posture data and the target posture data; Using the initial posture data and the target posture data to form a six-dimensional vector as sampling points, performing fracture reduction path planning to generate a final path; Optimizing the final path through a path post-processing algorithm; The step of using the initial posture data and the target posture data to form a six-dimensional vector as sampling points to perform fracture reduction path planning and generate a final path specifically includes the following steps: Use the initial position as the node of the tree and build a tree including the starting point; By iterating the enhanced RRT* algorithm, a node χ is randomly sampled from the state space. rand , and find the node X closest to this node in the tree nearest ; The node χ is transformed into rand Expand to a node with a step length χ new ; If we judge χ new and χ nearest If the line between them collides with an obstacle, the node is abandoned; For the nodes that are not abandoned, select a list of all neighboring nodes on the tree whose distance is less than the preset threshold X near , perform two optimization steps; The step of optimizing the final path by using a path post-processing algorithm specifically includes the following steps: Starting from the starting point, check whether the Lth point is directly connected to the starting point and there is no collision with obstacles on the path; if there is no collision, directly connect and delete the intermediate nodes; if there is a collision, check the L-1th point until a collision-free node is found; starting from this node, repeat the test until reaching the target point, and finally return the remaining nodes and edges; The path processed by the path short-circuiting algorithm retains the n+1 key nodes V that the broken bones must pass through during the reduction process. i , after connecting end to end, a geometrically continuous n-segment line is generated, corresponding to n intervals; Smooth the line segments connected by the key nodes to obtain a space curve that meets the parameter continuity conditions, where the parameter continuity includes geometric continuity, velocity continuity, and acceleration continuity; specifically: The cubic spline interpolation method is used to segmentally smooth the reset path. The segmented parametric curve in three-dimensional space is defined as: C(t)={C i (τ),τ∈[τ i ,t i+1 ],i=0,…,n-1} Where τ0=0, τ n =1; corresponding to the interpolation coordinates in each interval, the cubic spline interpolation definition is used: Among them, there are 12n unknown parameters in n intervals, and they are solved through 12n equations; The specific constraints include: the function values at the beginning and end and internal nodes are known, generating 6n equations; the first-order derivatives at adjacent interval nodes are equal, generating 3(n-1) equations; the second-order derivatives at adjacent interval nodes are equal, generating 3(n-1) equations; the second-order derivatives at the beginning and end of the entire path are zero, generating 6 equations; the formula is expressed as: The position coordinates V of the path nodes are i (p) and posture coordinates V i (θ) can be interpolated to obtain a smooth motion trajectory.

2. A robot-assisted fracture reduction path planning method according to claim 1, characterized in that: The step of segmenting and reconstructing the CT data and planning the target posture to obtain the initial posture data and the target posture data corresponding to the broken bone specifically includes the following steps: Segment and reconstruct the CT data to form a configuration space containing the broken bones The configuration space is divided into two categories: healthy bones and bones that do not need to be moved Defined as obstacle area, other areas It is defined as a non-obstruction area, and the initial posture data of the broken bone is defined as χ start ; Select the target posture data x for the broken bone target , restore the physiological state of the pelvic structure.

3. The robot-assisted fracture reduction path planning method according to claim 1, characterized in that: The step of calculating the position difference and posture difference between the initial pose data and the target pose data specifically includes the following steps: Let the Euler angle at the initial posture be θ s =(0,0,0), then the Euler angle of the target posture is θ t =(α t ,β t ,γ t ); The initial pose data and the target pose data are defined as vectors of length 6 respectively, then: Among them, P s and P t are the centroid coordinates of the point cloud at the initial and target states of the bone fragments, respectively; If at time τ and in the Cartesian coordinate system, the center of mass of the broken bone moves to (x τ ,y τ ,z τ ), the Euler angles are rotated by (α τ ,β τ ,γ τ ), then the initial position of the bone point cloud F v Expressed as: F τ =R Z (c τ )R Y (b τ )R X (a τ )·(F s -P s )+(x τ ,y τ ,z τ ), Among them, F s is the coordinate set of the initial state of the bone fragment point cloud; P s is the coordinate of the point cloud centroid of the initial state of the bone fragment; R X 、R Y 、R Z are the rotation matrices around the X, Y, and Z axes respectively.

4. The robot-assisted fracture reduction path planning method according to claim 1, characterized in that: For the nodes that are not abandoned, a list of all neighboring nodes X with a distance less than a preset threshold is selected on the tree. near , the two optimization steps specifically include the following steps: In the first optimization step, we select χ new The parent node of the node is calculated from the starting point through each neighboring node χ near Arrival new Cost; The neighboring node χ on the shortest cost path near Let χ new The parent node of new Add to the tree and become a node on the tree; In the second optimization step, the neighboring nodes χ are selected near The parent node of new Arrival at neighboring node χ near the price; If through χ new Arrival near The cost is less than the neighboring node χ near The original cost of χ near The parent node is set to χ new ; When the maximum number of iterations is reached or the end condition is met, a reverse search starts from the target point to generate the final path.

5. The robot-assisted fracture reduction path planning method according to claim 1, characterized in that: The enhanced RRT* algorithm also includes: The nodes are expanded by synchronously exploring asynchronous adjustments, and the position coordinates and Euler angles are adjusted separately with steps of Δ1 and Δ2 respectively. The collision detection algorithm based on the surface point cloud of the object is used to determine the moving bone F temp With other bone blocks F obs Whether there is a collision between them is determined by detecting whether there is an intersection S between the point clouds: S=F temp ∩F obs Traverse F temp For each point in F, calculate the obs The nearest neighbor points in ; If there are neighboring points within the preset threshold δ, it is determined that a collision occurs, otherwise it is determined that no collision occurs; The time complexity of the collision detection algorithm is O(n 2 ), KD tree is used to model the environmental obstacle point cloud, reducing the time complexity to O(nlogn).

6. A system for a robot-assisted fracture reduction path planning method according to claim 1, characterized in that: include: Acquisition module: acquires the patient's CT data; Data processing module: performs segmentation reconstruction and target posture planning on the CT data to obtain initial posture data and target posture data corresponding to the broken bone; Calculating a position difference and a posture difference between the initial posture data and the target posture data; Fracture reduction path planning module: using the initial posture data and the target posture data to form a six-dimensional vector as sampling points, performs fracture reduction path planning and generates a final path; Optimization module: optimizes the final path through a path post-processing algorithm.

7. A computer device, characterized in that: The computer device includes a processor and a memory coupled to the processor, wherein the memory stores program instructions for implementing a robot-assisted fracture reduction path planning method as described in any one of claims 1-5; the processor is used to execute the program instructions stored in the memory to implement a robot-assisted fracture reduction path planning.

8. A storage medium, characterized in that: Program instructions executable by a processor are stored, and the program instructions are used to execute the robot-assisted fracture reduction path planning method according to any one of claims 1 to 5.

Images