Flexible needle lung biopsy diameter planning method and related devices

Abstract

This invention discloses a flexible needle lung biopsy path planning method and related apparatus. The method includes: inputting a configuration space, a starting configuration, a target region, an obstacle region, and a preset branch angle, wherein the target region has a target deviation probability and an uncertainty range; generating a random configuration through a target bias sampling strategy, selecting random nodes with a certain probability and uncertainty range as target nodes to adapt to tumor puncture biopsy path planning at different locations; constructing a search tree to find the path from the starting configuration to the target region, wherein the branch angle of the search tree is limited to a preset branch angle range to ensure that there are no sharp corners or abrupt turns in the path. This invention improves the convergence speed of the algorithm by increasing the flexible needle parameters and the target deviation of the random point generation, enabling it to reach the target region faster, thereby better simulating the path of the flexible needle inside the tissue and minimizing tissue damage caused by large turning angles.

Description

Technical Field

[0001] This invention relates to the field of medical technology, specifically to a flexible needle lung biopsy diameter planning method and device, computing equipment, and computer program product. Background Technology

[0002] Current puncture techniques primarily utilize rigid needles combined with imaging techniques to perform biopsies on lesions in the patient's lungs. However, these techniques require patients to hold their breath during the biopsy, posing a challenge for patients with poor physical condition and the elderly. To address this difficulty, flexible needle puncture techniques have emerged. Because of their non-rigid nature, flexible needles can be guided to the lesion during surgery, eliminating the need for patients to hold their breath. Therefore, preoperative pathway planning for flexible needles becomes a crucial step in ensuring the success of the procedure.

[0003] Existing RRT*-based path planning methods do not address the parameters of flexible needles and target deviations from random point generation, resulting in slow convergence speeds, inability to reach the target area quickly, poor adaptation to the simulated human respiratory system environment, and often causing tissue damage due to their large turning angles.

[0004] Therefore, developing a flexible needle lung biopsy diameter planning method has become an urgent problem to be solved. Summary of the Invention

[0005] In view of the above problems, the present invention is proposed to provide a flexible needle lung biopsy diameter planning method and apparatus, computing device, and computer program product that overcomes the problem of tissue damage caused by large turning angles.

[0006] According to one aspect of the present invention, a method for planning the diameter of a flexible needle lung biopsy is provided, comprising:

[0007] Input configuration space, starting configuration, target area, obstacle area and preset branch angle, wherein the target area has target deviation probability and uncertainty range;

[0008] Random configurations are generated through a target-biased sampling strategy, and random nodes with a certain probability and uncertainty range are selected as target nodes to adapt to tumor puncture biopsy path planning at different locations.

[0009] A search tree is constructed to find a path from the initial configuration to the target area, wherein the branch angles of the search tree are limited to a range of preset branch angles to ensure that there are no sharp corners or turns in the path.

[0010] In one alternative approach, the search tree generation path incorporates a Bézier curve smoothing technique to generate a continuous and smooth final path.

[0011] In an alternative approach, the control points of the Bézier curve are determined based on the midpoints of line segments in the search path, and the initial and last points are connected to the midpoints of their respective nearest points.

[0012] In an alternative approach, the method further includes:

[0013] When connecting two adjacent Bézier curves, ensure that at the common intersection point, the termination control point vector of the preceding curve is the same as the starting control point vector of the following curve to achieve geometric continuity; and,

[0014] The tangent vectors have the same direction and magnitude to ensure that the derivative of the curve is continuous at the connection point, thus achieving tangent continuity.

[0015] In an alternative approach, the method further includes:

[0016] The search path is converted into a Bezier curve, and collision detection is performed on the new smooth path to ensure that the path does not intersect with the obstacle area and meets the limitation that the maximum turning angle does not exceed a preset angle.

[0017] In an alternative approach, the method further includes:

[0018] The smoothness of the curve is controlled by the degree of the Bézier curve, given n+1 control points P0, P1, P2, ..., P n A Bézier curve of degree n is defined as:

[0019]

[0020] Among them, B i，n (t) is called the Bezier basis function or Bernstein polynomial, where n is the number of control points.

[0021] In one alternative approach, the derivative of the Bézier curve is:

[0022]

[0023] in,

[0024] Q0=n(P1-P0),Q1=n(P2-P1),...,Q n-1 =n(P n -P n-1 ), Q i , t=0, ..., n-1, t=0 is the first control point, t=1 is the last control point.

[0025] According to another aspect of the present invention, a flexible needle lung biopsy diameter planning device is provided, comprising:

[0026] According to another aspect of the present invention, a computing device is provided, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus;

[0027] According to another aspect of the present invention, a computer program product is provided, comprising at least one executable instruction that causes the processor to perform operations corresponding to the above-described flexible needle lung biopsy diameter planning method.

[0028] According to the solution provided by the present invention, the method includes an input configuration space, a starting configuration, a target region, an obstacle region, and a preset branch angle. The target region has a target deviation probability and an uncertainty range. A random configuration is generated through a target bias sampling strategy, and random nodes with a certain probability and uncertainty range are selected as target nodes to adapt to tumor puncture biopsy path planning at different locations. A search tree is constructed to find the path from the starting configuration to the target region. The branch angle of the search tree is limited to the range of the preset branch angles to ensure that there are no sharp angles or turns in the path. The present invention improves the convergence speed of the algorithm by increasing the flexible needle parameters and the target deviation of the random point generation, enabling it to reach the target region faster, thereby better simulating the path of the flexible needle within the tissue and minimizing tissue damage caused by large turning angles.

[0029] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0030] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0031] Figure 1 A flowchart illustrating the flexible needle lung biopsy diameter planning method according to an embodiment of the present invention is shown.

[0032] Figure 2 A schematic diagram of the RRT* algorithm incorporating Bezier smoothing curves according to an embodiment of the present invention is shown;

[0033] Figure 3 A schematic diagram of the branch angle according to an embodiment of the present invention is shown;

[0034] Figure 4 A schematic diagram of a Bezier smoothing path according to an embodiment of the present invention is shown;

[0035] Figure 5 A schematic diagram comparing RRT and RRT* with Bezier curve smoothing according to an embodiment of the present invention is shown;

[0036] Figure 6 A comparative schematic diagram of the RRT of an embodiment of the present invention and the RRT* with Bezier curve smoothing is shown;

[0037] Figure 7 A schematic diagram of the frame of the flexible needle lung biopsy diameter planning device according to an embodiment of the present invention is shown;

[0038] Figure 8 A schematic diagram of the structure of a computing device according to an embodiment of the present invention is shown. Detailed Implementation

[0039] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0040] The Rapidly-exploring Random Trees Star (RRT) path planning algorithm is an asymptotically optimal path planning algorithm based on random sampling. Its basic principle is to start from an initial point, generate nodes in the state space through random sampling, and attempt to connect these nodes to a random tree, thus forming an ever-expanding random tree. When a leaf node in the random tree contains the target point or enters the target region, a path from the initial point to the target point is found by backtracking in the random tree. However, existing RRT* techniques cannot well adapt to the environment simulating the human respiratory system, and their convergence speed is relatively slow. This invention improves upon existing RRT* path planning by adding specific parameters for existing flexible needles and a goal-bias in random point generation. The goal-bias significantly improves the algorithm's convergence speed, enabling it to reach the target region faster. Furthermore, by adding flexible needle parameters (mainly by limiting the maximum number of turning angles and radius of the code), the algorithm better simulates the path of the flexible needle within the tissue, minimizing damage to the tissue caused by large turning angles. The invention is described in detail below with reference to specific embodiments.

[0041] Figure 1 A flowchart illustrating the flexible needle lung biopsy diameter planning method according to an embodiment of the present invention is shown. Specifically, as... Figure 1As shown, it includes the following steps:

[0042] Step S101: Input configuration space, starting configuration, target area, obstacle area and preset branch angle, wherein the target area has target deviation probability and uncertainty range.

[0043] In this embodiment, as Figure 2 As shown, the input parameters are configuration space, starting configuration, target area, obstacle area, and preset branch angle. The configuration space... Define the set of all possible states in which the flexible needle can move, which can be two-dimensional (e.g., moving on a plane), three-dimensional (e.g., moving in space), or higher-dimensional (including velocity, acceleration, etc.) space. The target region is defined as the target configuration. A spherical region centered on the area. Initial configuration. This is the starting point for path planning, i.e., the position where the flexible needle begins to move. Target area. The goal of path planning is the desired location for the flexible needle. The target area has a target deviation probability and uncertainty range; that is, the target area is a region with a certain probability distribution, requiring attempts to reach multiple points within this area, or path optimization to maximize the probability of reaching the target area. Obstacle area. Areas where flexible needles cannot enter or pass through; obstacle areas are used to avoid collisions when expanding the tree. Preset branch angle. Limiting the range of directions in which new nodes can connect to existing nodes when a random tree expands helps control the growth direction of the tree, preventing it from becoming too cluttered or failing to cover certain important areas. At the same time, it also reflects the limitations of the flexible needle's movement capabilities (such as the maximum turning angle).

[0044] Step S102: A random configuration is generated through a target-biased sampling strategy, and a random node with a certain probability and uncertainty range is selected as the target node to adapt to the tumor puncture biopsy path planning at different locations.

[0045] Path planning assumes that the kinematic model of the CCS zigzag needle is valid during trajectory turning. In traditional RRT planning, the path is a series of continuous arcs tangent to each other. The constant curvature of each arc must not exceed the maximum natural curvature. Improved RRT with arc generation by a local planner can search for multiple feasible paths, but the generated paths are often long due to redundant distortions. To address these issues, Klarman et al. proposed the RRT* path planning algorithm, which removes redundant edges, thus significantly reducing the total path length. In this embodiment, the original RRT* is modified for lung biopsy. One of the modifications is the introduction of target bias, allowing the target node to be selected as a random node with a certain probability and uncertainty range. This is to accommodate changes in tumor location due to lung movement and results in faster convergence.

[0046] Step S103: Construct a search tree to find a path from the initial configuration to the target area, wherein the branch angle of the search tree is limited to the range of the preset branch angle to ensure that there are no sharp corners or sharp turns in the path.

[0047] like Figure 2 As shown, a search tree is constructed to find the path from the initial configuration to the target region. In this embodiment, another modification to the original RRT* for lung biopsy is to limit the branch angles of the search tree to a range of preset branch angles, such as... Figure 3 As shown, this is to ensure that there are no sharp corners or turns along the created path, thereby reducing tissue damage.

[0048] like Figure 4 As shown, the search tree generation path incorporates Bézier curve smoothing technology to generate a continuous and smooth final path. The control points of the Bézier curve are determined based on the midpoints of line segments in the search path, and the initial and last points are connected to the midpoints of their respective nearest points. When connecting two adjacent Bézier curves, it is ensured that at the common intersection, the termination control point vector of the preceding curve is the same as the starting control point vector of the following curve to achieve geometric continuity (G1 continuity); and that the direction and magnitude of the tangent vectors are the same to ensure that the derivatives of the curves are continuous at the connection point to achieve tangent continuity (C1 continuity). This embodiment considers smooth path evolution in the path planning problem without sharp turns, not only for practical convenience but also from the perspective of minimizing tissue damage. The maximum turning angle is specified to not exceed a preset branch angle (10 degrees). The smoothing process involves finding the midpoint between the second edge and the penultimate edge of path Ψ. The Bézier curve is established by connecting consecutive midpoints (except for the initial and last points), where these two points are connected to the midpoint of the nearest point. Since there are three non-collinear control points, each is a quadratic curve lying in the plane. Therefore, the continuity of G1 and C1 is implicitly achieved by connecting the continuous Bézier curves at their common intersection points, where the connection vector on each respective Bézier curve (denoted as C) is... m-1 C m It is the same as D0D1. After smoothing, a new path (blue curve) is generated along the trajectory of the Bézier curve connecting the start and end points.

[0049] In an optional implementation, the method further includes:

[0050] The search path is converted into a Bezier curve, and collision detection is performed on the new smooth path to ensure that the path does not intersect with the obstacle area and meets the limitation that the maximum turning angle does not exceed a preset angle.

[0051] In this embodiment, the RRT* algorithm is used to generate a path from the initial configuration to the target area. This path consists of a series of nodes, which may not be perfectly smooth. These path points are used as control points for a Bézier curve to generate a smooth Bézier curve. Depending on the desired smoothness, a second-, third-, or higher-order Bézier curve can be used. After converting the path to a Bézier curve, collision detection is performed to ensure that the newly generated smooth path does not intersect with obstacle areas. Collision detection involves sampling points on the Bézier curve and checking if these points are located within obstacle areas. If any sampled point is found to be within an obstacle area, it indicates that the path intersects with an obstacle. If a turning angle at a certain location exceeds a preset limit, the turning angle at that location is reduced by adjusting the control points of the Bézier curve.

[0052] In an optional implementation, the method further includes:

[0053] The smoothness of the curve is controlled by the degree of the Bézier curve, given n+1 control points P0, P1, P2, ..., P n A Bézier curve of degree n is defined as:

[0054]

[0055] Among them, B i，n (t) is called the Bezier basis function or Bernstein polynomial, where n is the number of control points. The function is continuous and differentiable, therefore its corresponding function curve, i.e. the smoothed curve, must be smooth.

[0056] Differentiating the parametric function of the Bézier curve yields:

[0057]

[0058] set up

[0059] Q0=n(P1-P0), Q1=n(P2-P1),...,Q n-1 =n(P n -P n-1 If ), then P′(t) is:

[0060]

[0061] in,

[0062] Q0=n(P1-P0), Q1=n(P2-P1),...,Q n-1 =n(P n -P n-1 ), Q i, t=0, ..., n-1, t=0 is the first control point, t=1 is the last control point.

[0063] Therefore, the first derivative of an n-degree Bézier curve is an n-1 degree Bézier curve. This property of Bézier curves allows for smoothing of jagged, segmented paths. Thus, the curve passes through the first and last control points t=0 and t=1, respectively. P′(0) = n(P1-P0) and P′(1) = n(P1-P0). n -P n-1 The directions of the first and last tangent vectors of the Bézier curve are represented by P0 and P1, respectively. n-1 P n The first and last line segments are defined. Therefore, G1 continuity naturally occurs when two Bézier curves (e.g., C(t) and D(t) defined by m+1 control points) are connected at a common point, because C... m =D0 is from C m-1 C m The intersection point and direction defined by D0D1 are in the same direction. To achieve C1 continuity, C... m-1 C m It must have the same amplitude as D0D1.

[0064] Next, combined Figure 2 The algorithm shown explains in principle why the above improvements achieve the desired effect. The input of the algorithm is the allocation space, using... This indicates the initial configuration. target area and obstacle area target area Defined as target configuration x goal The search tree is a spherical region centered on V. V and E represent the vertices and edges of the search tree, respectively. The search tree is first initialized with the contained V and E is set to empty. The tree branches progressively by generating random nodes using a target-biased sampling strategy. If there is no target bias, completely random nodes x are drawn from free space according to a uniform distribution. rand Under the target biased sampling method, the target node x goal The nodes are set as random nodes with a certain probability within the target area. Then, each node in the tree is searched and traversed to find the node closest to x based on Euclidean distance. rand node x nearest If x nearest and x rand If the distance between them is greater than one step, then according to x near A new node x is generated in the specified direction with a step size. newest and x rand Otherwise, x randSet to x new Next, a collision check is performed to ensure x new Not located middle. Figure 5 The function get_near_nodes() in x allows you to... new Local search for a set of potential parent nodes x within the defined region new Parent node x min It is based on the standard of minimum distance from the state node and within the "branch angle" range from x new The selected node. After selecting the parent node, x... min and x new Connect with straight lines. Then perform tree rewiring. When x near exist When the above process ends, the path Ψ is traced and extracted. Otherwise, the iteration is repeated until x is found. new exist middle.

[0065] The method of this embodiment is compared with an RRT algorithm based on simulation using a 2D environment with randomly set obstacles. The configuration space is selected as... Table 1 shows the initial configuration. and target configuration x goal Two different settings.

[0066] Table 1: Setting the start and end points

[0067] test <![CDATA[x start ]]> <![CDATA[x goal ]]> 1 (-60，20) (95，160) 2 (80，20) (-90，160)

[0068] The specific results of the experiment, such as Figure 5 , Figure 6 As shown. In Figure 5 In the diagram, (a) the search tree of the RRT. (b) the final path of the RRT. (c) the RRT* tree smoothed using a Bezier curve. (d) the RRT* path before smoothing. (e) the final path of the RRT* smoothed using a Bezier curve. Figure 6 In the diagram, (a) the search tree of the RRT. (b) the final path of the RRT. (c) the RRT* tree smoothed using a Bezier curve. (d) the RRT* path before smoothing. (e) the final path of the RRT* smoothed using a Bezier curve.

[0069] To compare the performance of the two methods, 30 simulations were performed for each setting. The average search time, success rate, average total length, and average number of twists to achieve a successful solution were compared. The number of twists corresponds to the number of times the needle needed to be reoriented. The comparison results are shown in Table 2. Based on the results of different starting and target configuration settings, the method in this embodiment outperforms RRT and, on average, calculates a shorter, less tortuous path. However, RRT* using Bézier curve smoothing takes longer to compute than RRT. This is primarily due to the angle constraints during tree expansion and the parametric smoothing of the Bézier curve. Physicians can strike a balance between a shorter, smoother path with slightly longer computation time and a longer, more tortuous path. In real-world situations where tissue healing is difficult and tissue damage is considered fatal, the path planned by the method in this embodiment would be a better choice.

[0070] Table 2: Comparison results of Experiment 1 and Experiment 2

[0071]

[0072] According to the solution provided by the present invention, the algorithm includes an input configuration space, a starting configuration, a target region, an obstacle region, and a preset branch angle. The target region has a target deviation probability and an uncertainty range. A random configuration is generated through a target bias sampling strategy, and random nodes with a certain probability and uncertainty range are selected as target nodes to adapt to tumor puncture biopsy path planning at different locations. A search tree is constructed to find the path from the starting configuration to the target region. The branch angle of the search tree is limited to the range of the preset branch angle to ensure that there are no sharp angles or turns in the path. The present invention improves the convergence speed of the algorithm by increasing the flexible needle parameters and the target deviation of the randomly generated points, enabling it to reach the target region faster, thereby better simulating the path of the flexible needle within the tissue and minimizing tissue damage caused by large turning angles.

[0073] Figure 7 A schematic diagram of the flexible needle lung biopsy path planning device according to an embodiment of the present invention is shown. The flexible needle lung biopsy path planning device includes: an input module 710, a target node selection module 720, and a search tree construction module 730.

[0074] The input module 710 is used to input the configuration space, the starting configuration, the target area, the obstacle area, and the preset branch angle, wherein the target area has a target deviation probability and an uncertainty range;

[0075] The target node selection module 720 is used to generate a random configuration through a target bias sampling strategy, and select a random node with a certain probability and uncertainty range as the target node to adapt to the tumor puncture biopsy path planning at different locations.

[0076] The search tree construction module 730 is used to construct a search tree to find a path from the starting configuration to the target area, wherein the branch angle of the search tree is limited to the range of the preset branch angle to ensure that there are no sharp corners or sharp turns in the path.

[0077] In one alternative approach, the search tree generation path incorporates a Bézier curve smoothing technique to generate a continuous and smooth final path.

[0078] In an alternative approach, the control points of the Bézier curve are determined based on the midpoints of line segments in the search path, and the initial and last points are connected to the midpoints of their respective nearest points.

[0079] In an alternative approach, the method further includes:

[0080] When connecting two adjacent Bézier curves, ensure that at the common intersection point, the termination control point vector of the preceding curve is the same as the starting control point vector of the following curve to achieve geometric continuity; and,

[0081] The tangent vectors have the same direction and magnitude to ensure that the derivative of the curve is continuous at the connection point, thus achieving tangent continuity.

[0082] In an alternative approach, the method further includes:

[0083] The search path is converted into a Bezier curve, and collision detection is performed on the new smooth path to ensure that the path does not intersect with the obstacle area and meets the limitation that the maximum turning angle does not exceed a preset angle.

[0084] In an alternative approach, the method further includes:

[0085] The smoothness of the curve is controlled by the degree of the Bézier curve, given n+1 control points P0, P1, P2, ..., P n A Bézier curve of degree n is defined as:

[0086]

[0087] Among them, B i，n (t) is called the Bezier basis function or Bernstein polynomial, where n is the number of control points.

[0088] In one alternative approach, the derivative of the Bézier curve is:

[0089]

[0090] in,

[0091] Q0=n(P1-P0), Q1=n(P2-P1),...,Q n-1 =n(P n -P n-1 ), Q i , t=0, ..., n-1, t=0 is the first control point, t=1 is the last control point.

[0092] Figure 8 The diagram shows a structural schematic of an embodiment of the computing device of the present invention. The specific embodiments of the present invention do not limit the specific implementation of the computing device.

[0093] like Figure 8 As shown, the computing device may include: a processor 802, a communications interface 804, a memory 806, and a communications bus 808.

[0094] The processor 802, communication interface 804, and memory 806 communicate with each other via communication bus 808. Communication interface 804 is used to communicate with other network elements such as clients or other servers. Processor 802 executes program 810, specifically performing the relevant steps in the above-described embodiment of the flexible needle lung biopsy diameter planning method.

[0095] Specifically, program 810 may include program code that includes computer operation instructions.

[0096] Processor 802 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The computing device includes one or more processors, which may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0097] Memory 806 is used to store program 810. Memory 806 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0098] According to the solution provided by the present invention, the algorithm includes an input configuration space, a starting configuration, a target region, an obstacle region, and a preset branch angle. The target region has a target deviation probability and an uncertainty range. A random configuration is generated through a target bias sampling strategy, and random nodes with a certain probability and uncertainty range are selected as target nodes to adapt to tumor puncture biopsy path planning at different locations. A search tree is constructed to find the path from the starting configuration to the target region. The branch angle of the search tree is limited to the range of the preset branch angle to ensure that there are no sharp angles or turns in the path. The present invention improves the convergence speed of the algorithm by increasing the flexible needle parameters and the target deviation of the randomly generated points, enabling it to reach the target region faster, thereby better simulating the path of the flexible needle within the tissue and minimizing tissue damage caused by large turning angles.

[0099] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination of all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or device so disclosed can be employed. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose. Furthermore, those skilled in the art will understand that although some embodiments herein include certain features included in other embodiments but not others, combinations of features from different embodiments are intended to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments can be used in any combination. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims listing several devices, several of these devices may be embodied by the same hardware item. Unless otherwise specified, the steps in the above embodiments should not be construed as limiting the order of execution.

Claims

1. A method for planning the diameter of a flexible needle lung biopsy, characterized in that, include: Input configuration space, starting configuration, target area, obstacle area and preset branch angle, wherein the target area has target deviation probability and uncertainty range; Random configurations are generated through a target-biased sampling strategy, and random nodes with a certain probability and uncertainty range are selected as target nodes to adapt to tumor puncture biopsy path planning at different locations. A search tree is constructed to find a path from the initial configuration to the target area, wherein the branch angle of the search tree is limited to the range of the preset branch angle to ensure that there are no sharp corners or sharp turns in the path; The search tree generation path incorporates Bézier curve smoothing technology to generate a continuous and smooth final path. The control points of the Bézier curve are determined based on the midpoints of the line segments in the search path, and the initial point and the last point are connected to the midpoint of their respective nearest points; When connecting two adjacent Bézier curves, ensure that at the common intersection point, the termination control point vector of the preceding curve is the same as the starting control point vector of the following curve to achieve geometric continuity; and, The tangent vectors have the same direction and magnitude to ensure that the derivative of the curve is continuous at the connection point, thus achieving tangent continuity; The search path is converted into a Bezier curve, and collision detection is performed on the new smooth path to ensure that the path does not intersect with the obstacle area and meets the limitation that the maximum turning angle does not exceed a preset angle.

2. The flexible needle lung biopsy diameter planning method according to claim 1, characterized in that, The method further includes: The smoothness of the curve is controlled by the degree of the Bézier curve, given n+1 control points. A Bézier curve of degree n is defined as: in, These are called Bezier basis functions or Bernstein polynomials, where n is the number of control points. .

3. The flexible needle lung biopsy diameter planning method according to claim 2, characterized in that, The derivative of the Bézier curve is: in, , t=0 is the first control point, and t=1 is the last control point.

4. A flexible needle lung biopsy diameter planning device, characterized in that, include: The input module is used to input the configuration space, starting configuration, target area, obstacle area and preset branch angle, wherein the target area has a target deviation probability and uncertainty range; The target node selection module is used to generate a random configuration through a target bias sampling strategy, and select random nodes with a certain probability and uncertainty range as target nodes to adapt to tumor puncture biopsy path planning at different locations. A search tree construction module is used to construct a search tree to find a path from the starting configuration to the target area, wherein the branch angle of the search tree is limited to the range of the preset branch angle to ensure that there are no sharp corners or sharp turns in the path; The search tree generation path incorporates Bézier curve smoothing technology to generate a continuous and smooth final path. The control points of the Bézier curve are determined based on the midpoints of the line segments in the search path, and the initial point and the last point are connected to the midpoint of their respective nearest points; When connecting two adjacent Bézier curves, ensure that at the common intersection point, the termination control point vector of the preceding curve is the same as the starting control point vector of the following curve to achieve geometric continuity; and, The tangent vectors have the same direction and magnitude to ensure that the derivative of the curve is continuous at the connection point, thus achieving tangent continuity; The search path is converted into a Bezier curve, and collision detection is performed on the new smooth path to ensure that the path does not intersect with the obstacle area and meets the limitation that the maximum turning angle does not exceed a preset angle.

5. A computing device, comprising: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction that causes the processor to perform the operation corresponding to the flexible needle lung biopsy diameter planning method as described in any one of claims 1-3.

6. A computer program product comprising at least one executable instruction that causes a processor to perform an operation corresponding to the flexible needle lung biopsy path planning method as described in any one of claims 1-3.

Images