Interventional procedure pathway planning systems, methods, apparatuses, and storage media

Abstract

The application relates to an interventional operation path planning system, method, device and storage medium. The system comprises a processor which performs the following: acquiring an initial needle entry path of an interventional operation, the initial needle entry path comprising a target point and an initial needle entry point; generating a candidate needle entry point based on the initial needle entry point, generating at least one neighborhood candidate path based on the target point and the candidate needle entry point; and determining a target path based on the neighborhood candidate path, a lightweight constraint of a mechanical arm joint angle and a collision constraint. The system can generate a candidate point based on the initial needle entry point, generate at least one neighborhood candidate path covering the neighborhood space of the initial needle entry point, break through the limitation of a single path, provide a basis for searching for a target path in the neighborhood candidate path, and evaluate the neighborhood candidate path in combination with the lightweight constraint of the mechanical arm joint angle and the collision constraint, so that the target path is accurately and efficiently obtained in multiple dimensions, and the planning quality and efficiency of robot-assisted puncture are systematically improved.

Description

Technical Field

[0001] This application relates to the field of minimally invasive surgery technology, and in particular to an interventional surgical path planning system, method, device and storage medium. Background Technology

[0002] With the development of modern minimally invasive surgical techniques, robot-assisted puncture technology has been widely used due to its high precision and stability. However, in actual operation, the path planning of the robotic arm is often limited by a variety of factors, such as the complexity of human anatomy, the range of motion of the robotic arm itself, and the potential risk of collision with the surrounding environment. Traditional path planning methods usually only focus on the feasibility analysis of a single path, lacking a comprehensive assessment of the feasibility within the path's neighborhood. This may lead to inefficiency for users when adjusting the path or difficulty in finding a surgical path that the robotic arm can execute. Summary of the Invention

[0003] Therefore, it is necessary to provide an interventional surgical path planning system, method, device, and storage medium to address the aforementioned technical problems.

[0004] An interventional surgical pathway planning system, the system including a processor, the processor being configured to perform the following steps:

[0005] Obtain the initial needle insertion path for interventional surgery, wherein the initial needle insertion path includes the target point and the initial needle insertion point;

[0006] Based on the initial needle insertion point, candidate needle insertion points are generated, and based on the target point and the candidate needle insertion points, at least one neighborhood candidate path is generated.

[0007] The target path is determined based on the neighborhood candidate paths, the lightweight constraints of the robot arm joint angles, and the collision constraints.

[0008] In one embodiment, the processor's execution of generating candidate needle insertion points based on the initial needle insertion point, and generating at least one neighborhood candidate path based on the target point and the candidate needle insertion points, further includes performing the following steps: obtaining a plane passing through the initial needle insertion point and perpendicular to the initial needle insertion path; determining a candidate region on the plane based on the initial needle insertion point and a preset range; generating discrete points on the candidate region; generating the candidate needle insertion point based on the discrete points; and generating at least one neighborhood candidate path based on the target point and the candidate needle insertion points.

[0009] In one embodiment, the processor's execution of determining the target path based on the neighborhood candidate path, lightweight constraints on the robot arm joint angles, and collision constraints further includes performing the following steps: constructing a first pose error matrix based on a first pose, a second pose, and a first joint angle; solving for the second joint angle based on the first pose error matrix; the first pose is used to simulate the pose of the robot arm end effector at the initial iteration point; the first joint angle is used to simulate the joint angles of the robot arm when the end effector is in the first pose; the second pose is used to simulate the pose of the robot arm end effector when it is located at the candidate needle insertion point of the neighborhood candidate path; and the second joint angle is used to simulate the... The process involves determining the joint angles of the robotic arm when its end effector is in the second pose; judging whether the neighboring candidate paths satisfy the lightweight constraint on the robotic arm joint angles; if the lightweight constraint is satisfied, the second joint angle has a solution when the robotic arm end effector is in the second pose; if the lightweight constraint is not satisfied, the second joint angle has no solution; if the second joint angle has a solution, judging whether the collision constraint is satisfied when the robotic arm end effector is in the second pose; and selecting the neighboring candidate paths that satisfy both the lightweight constraint on the robotic arm joint angles and the collision constraint as the target paths, where there are one or more target paths.

[0010] In one embodiment, the processor is further configured to perform the following steps: constructing a second pose error matrix based on a first pose, a third pose, and a first joint angle; solving for a third joint angle based on the second pose error matrix; the third pose is used to simulate the pose of the robotic arm end effector when it is located at the initial needle insertion point of the initial needle insertion path; the third joint angle is used to simulate the joint angle of the robotic arm when it is located in the third pose; when the robotic arm end effector is located in the third pose, the third joint angle has a solution; the third joint angle is used as an initial solution and introduced into the first pose error matrix to calculate the second joint angle.

[0011] In one embodiment, there are multiple neighborhood candidate paths, multiple candidate needle entry points for the neighborhood candidate paths, multiple second poses for the robotic arm end effector, and multiple first pose error matrices. The first pose error matrix corresponds one-to-one with the neighborhood candidate paths. The processor is further configured to perform the following steps: traversing and determining whether each neighborhood candidate path satisfies the lightweight constraint of the robotic arm joint angle; when any of the neighborhood candidate paths satisfies the lightweight constraint of the robotic arm joint angle, the robotic arm end effector is located in a sub-pose, the second pose includes the sub-pose, the sub-joint angle of the sub-pose error matrix has a solution, the first pose error matrix includes the sub-pose error matrix, and the second joint angle includes the sub-joint angle; using the sub-joint angle as an initial solution to introduce other first pose error matrices to calculate the corresponding second joint angle, where the other first pose error matrices are the first pose error matrices corresponding to the untraversed neighborhood candidate paths.

[0012] In one embodiment, the processor is further configured to perform the following steps: obtaining the target needle insertion point of the target path; determining the target needle insertion area based on the target needle insertion point; selecting an intervention needle insertion point within the target needle insertion area; and determining an intervention path based on the intervention needle insertion point and the target point, wherein the intervention path is one or more.

[0013] In one embodiment, the processor is further configured to perform the following steps: representing the target needle insertion region and the non-target needle insertion region with different colors, wherein the neighborhood candidate path is composed of the target path and the non-target path, and the non-target needle insertion region is composed of the needle insertion points of the non-target path.

[0014] In one embodiment, the processor is further configured to perform the following steps: adjusting the initial needle insertion path; the target needle insertion area and the non-target needle insertion area change as the initial needle insertion path is adjusted.

[0015] An interventional surgery path planning method includes: obtaining an initial needle insertion path for the interventional surgery, the initial needle insertion path including a target point and an initial needle insertion point; generating candidate needle insertion points based on the initial needle insertion point; generating at least one neighborhood candidate path based on the target point and the candidate needle insertion points; and determining a target path based on the neighborhood candidate paths, lightweight constraints on the joint angles of the robotic arm, and collision constraints.

[0016] An interventional surgery path planning device includes: an initial needle insertion path acquisition module for acquiring an initial needle insertion path for interventional surgery, the initial needle insertion path including a target point and an initial needle insertion point; a candidate needle insertion point generation module for generating candidate needle insertion points based on the initial needle insertion point; a neighborhood candidate path generation module for generating at least one neighborhood candidate path based on the target point and the candidate needle insertion point; and a target path determination module for determining a target path based on the neighborhood candidate paths, lightweight constraints on the joint angles of the robotic arm, and collision constraints.

[0017] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program performing the following steps:

[0018] Obtain the initial needle insertion path for interventional surgery, wherein the initial needle insertion path includes the target point and the initial needle insertion point;

[0019] Candidate needle entry points are generated based on the initial needle entry point. At least one neighborhood candidate path is generated based on the target point and the candidate needle entry points. The target path is determined based on the neighborhood candidate path, the lightweight constraint of the robot arm joint angle, and the collision constraint.

[0020] A computer-readable storage medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor: obtaining an initial needle insertion path for an interventional procedure, the initial needle insertion path including a target point and an initial needle insertion point; generating candidate needle insertion points based on the initial needle insertion point; generating at least one neighborhood candidate path based on the target point and the candidate needle insertion points; and determining a target path based on the neighborhood candidate paths, lightweight constraints on the joint angles of the robotic arm, and collision constraints.

[0021] Compared with the prior art, the present invention provides an interventional surgical path planning system, method, device, and storage medium, which have the following beneficial effects:

[0022] Candidate points are generated based on the initial needle insertion point, and at least one neighborhood candidate path is generated covering the neighborhood space of the initial needle insertion point. This breaks through the limitation of a single path and provides a basis for searching for the target path among the neighborhood candidate paths. Furthermore, the neighborhood candidate paths are evaluated by combining the lightweight constraints of the robotic arm joint angles and collision constraints. The multi-dimensional evaluation accurately and efficiently obtains the target path, thereby systematically improving the planning quality and efficiency of robot-assisted puncture. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of an interventional surgical path planning system in one embodiment;

[0024] Figure 2 This is a schematic diagram of the execution flow of an interventional surgical path planning system in one embodiment;

[0025] Figure 3 This is a schematic diagram of the process for generating neighborhood candidate paths in one embodiment;

[0026] Figure 4 This is a structural block diagram of an interventional surgical path planning device in one embodiment;

[0027] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0029] Figure 1 This is a schematic diagram of an interventional surgery path planning system according to an embodiment of this application. The terminal 102 communicates with the server 104 via a network. The server includes a processor and a memory. The memory stores at least one of a lightweight constraint model of the robotic arm joint angles, a full constraint model of the robotic arm joint angles, a collision constraint model, and an updated collision constraint model. The user can indicate the initial needle insertion path for the interventional surgery through the terminal 102. The initial needle insertion path includes a target point and an initial needle insertion point. The processor acquires the initial needle insertion path, the target point, and the initial needle insertion point. The processor generates candidate needle insertion points based on the initial needle insertion point. Further, the processor generates at least one neighborhood candidate path based on the target point and the candidate needle insertion points. The processor determines the target path based on the neighborhood candidate paths, the lightweight constraints of the robotic arm joint angles, and the collision constraints. The terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0030] In one embodiment, such as Figure 2 The flowchart 200 provided illustrates an execution flow for an interventional surgical path planning system. The system includes a processor 104, which performs the following steps:

[0031] Step 210: Obtain the initial needle insertion path for the interventional procedure. The initial needle insertion path includes the target point and the initial needle insertion point.

[0032] Step 220: Generate candidate needle entry points based on the initial needle entry point, and generate at least one neighborhood candidate path based on the target point and candidate needle entry points.

[0033] Step 230: Determine the target path based on neighborhood candidate paths, lightweight constraints on robot arm joint angles, and collision constraints.

[0034] Users can indicate the initial needle insertion path through interactive operations on the display interface. For example, users can manually point out the insertion point and target point on medical images based on experience, and connect the insertion point and target point to form the initial needle insertion path. The initial needle insertion path can also be automatically calculated by a computer based on medical images and clinical constraints. Medical images include at least one of clinical computed tomography (CT) images, magnetic resonance imaging (MRI) images, or ultrasound images. Clinical constraints include at least one of the target point, intervention depth, needle insertion direction, needle insertion angle, and distance from the path to the hazardous tissue. The initial needle insertion path includes the target point and the initial needle insertion point. The target point refers to the target point on the lesion during the interventional procedure, and the initial needle insertion point is the intersection point of the initial needle insertion path and the patient model's skin surface. For example, in a lung puncture interventional procedure, the target point is the target point on the lung lesion, and the insertion point is the intersection point of the path and the patient model's skin surface. The initial needle insertion point can be understood to include both location and orientation.

[0035] Candidate needle entry points are generated in the neighborhood of the initial needle entry point. One or more neighborhood candidate paths are generated based on the target point and the candidate needle entry points. The candidate needle entry points include two layers of connotation: position and attitude. When the neighborhood candidate path satisfies the lightweight constraint of the robot arm joint angle and the collision constraint, it can be determined as the target path.

[0036] Both offline and online interventional surgeries require robotic arm assistance for path execution. Offline interventional surgery can be understood as acquiring patient imaging data preoperatively, but not requiring real-time guidance from this data during the procedure. Online interventional surgery, on the other hand, requires patient imaging data for guidance during the procedure. The robotic arm has an end effector, on which an interventional instrument is attached. The movement of the robotic arm drives the movement of both the end effector and the instrument. Therefore, path planning for interventional surgeries needs to consider whether the robotic arm's end effector can reach candidate needle insertion points in the neighboring candidate paths. This can be referred to as robotic arm reachability, and satisfying the lightweight constraint on the robotic arm's joint angles means that the robotic arm is reachable. The lightweight constraint on the robotic arm's joint angles characterizes whether the solution for the joint angles of the robotic arm is within the range of joint rotation when the end effector moves to the target pose, i.e., the joint angles have a solution. The target pose can be understood as the candidate needle insertion point.

[0037] During the execution of offline or online interventional surgeries, the surgical environment includes imaging equipment and the patient. Therefore, the path planning for interventional surgeries needs to take into account the imaging equipment model and the patient model to avoid collisions and interference between the robotic arm and its end effector. Collision constraints are used to characterize that the robotic arm and its end effector do not interfere with the imaging equipment model during movement, and that the robotic arm and its end effector do not interfere with the patient model. Collision constraints can also be used to characterize that when the robotic arm's end effector reaches the target pose, it does not interfere with the imaging equipment model, and that the robotic arm and its end effector do not interfere with the patient model. The target pose can be understood as the candidate needle insertion point.

[0038] When the candidate path in the neighborhood is tilted, in order to keep the candidate path in the collimation range of the imaging equipment, the gantry of the imaging equipment needs to be tilted. The corresponding position of the patient in the bed needs to be changed. Therefore, the imaging equipment model and the patient model need to be updated. The updated model is introduced when judging whether the collision constraint is satisfied.

[0039] In the above system, candidate points are generated based on the initial needle insertion point, and at least one neighborhood candidate path is generated covering the neighborhood space of the initial needle insertion point. This breaks through the limitation of a single path and provides a basis for searching for the target path among the neighborhood candidate paths. Furthermore, the neighborhood candidate paths are evaluated by combining the lightweight constraints of the robotic arm joint angles and collision constraints. The multi-dimensional evaluation accurately and efficiently obtains the target path, thereby systematically improving the planning quality and efficiency of robot-assisted puncture.

[0040] In one embodiment, the robotic arm is an articulated robotic arm, where the movement of each joint is rotation around the joint axis; it has multiple degrees of freedom; for example, a five-degree-of-freedom robotic arm or a six-degree-of-freedom robotic arm. Each joint of the robotic arm corresponds to a joint angle, which refers to the angle of rotation of the corresponding joint around the joint axis. By changing the joint angle of each joint, the position or posture of the robotic arm's end effector can be changed.

[0041] A robotic arm has an end effector. When the robotic arm moves, it drives the end effector to move as well. When the end effector moves from an initial pose A to a target pose B, it can move directly from pose A to pose B; alternatively, one or more intermediate poses can be added between poses A and B. For example, the end effector moves from pose A to pose C, and then from pose C to pose B. In this case, the joints of the robotic arm need to rotate a certain angle around their joint axes to achieve the pose change of the end effector. Full constraint on the joint angles of the robotic arm can be understood as ensuring that the robotic arm is reachable not only in pose B but also in intermediate poses such as pose C. Light constraint on the joint angles of the robotic arm can be understood as not needing to consider intermediate poses; the robotic arm only needs to be reachable in pose B. Adding intermediate poses to the robotic arm's motion can be understood as breaking down a segment of the robotic arm's motion path into multiple segments, which can reduce the risk of collisions during the robotic arm's movement, but will increase the computational load. Full constraints on the joint angles of a robotic arm can be applied to a single path planning, such as the planning of the initial needle insertion path. Lightweight constraints on the joint angles of a robotic arm can be applied to multiple path planning, such as the planning of candidate paths in the neighborhood. Lightweight constraints on the joint angles of a robotic arm can be seen as a simplified calculation of full constraints on the joint angles of a robotic arm, which greatly reduces the amount of computation for multiple path planning.

[0042] In one embodiment, the initial needle insertion path is determined to meet the requirements of interventional surgical path planning based on the initial needle insertion path, the full constraint of the robotic arm joint angles, and the collision constraint. There are one or more intermediate points between the robotic arm and its end effector moving from the starting point to the initial needle insertion point. The joint angles of the robotic arm have solutions at both the intermediate points and the initial needle insertion point. Furthermore, during the movement, the robotic arm and its end effector do not interfere with the imaging device model or the patient model. When the end effector reaches the intermediate point, it does not interfere with the imaging device model or the patient model. When the end effector reaches the initial needle insertion point, it does not interfere with the imaging device model or the patient model. At this point, the initial needle insertion path is considered to meet the requirements of interventional surgical path planning, and it is used as the interventional path. The robotic arm assists in performing the interventional surgical operation, and steps 220-240 are not required. If the initial needle insertion path does not meet the full constraint of the robotic arm joint angle or does not meet the collision constraint, then it is considered that the initial needle insertion path does not meet the requirements of interventional surgery path planning, and steps 220-240 are executed.

[0043] In one embodiment, data acquisition and analysis can be performed before step 210. The processor acquires at least one medical image data from the patient's computed tomography (CT) scan, magnetic resonance imaging (MRI), or ultrasound images, and uses intelligent image processing technology to achieve accurate segmentation and 3D reconstruction of human tissue structures. First, the raw DICOM (Digital Imaging and Communications in Medicine) data is preprocessed, including grayscale normalization, isotropic resampling, and noise reduction enhancement. In the image segmentation stage, the system integrates deep learning methods with traditional algorithms, achieving progressive segmentation from coarse to fine through a cascaded segmentation framework. Deep learning methods include U-Net (U-shaped Convolutional Neural Network), nnUNet (No New-Net U-Net Framework), or DeepLab (Deep Labelling Network). Traditional algorithms include adaptive threshold segmentation, region growing, or level set evolution. For key anatomical structures, the system supports manual correction and contour delineation by users through interactive tools to ensure the clinical reliability of the segmentation results. For lesion target identification, the system employs a multi-scale feature analysis algorithm, combined with a prior knowledge base, to perform pathological feature matching, ultimately outputting structured data containing spatial coordinates, volume parameters, and tissue characteristics. All segmentation results are validated through a 3D visualization interface, generating obstacle maps and 3D models of surgical no-go zones required for robot motion planning, providing high-precision spatial constraints for subsequent path planning. The system also features dynamic update capabilities, enabling real-time fusion of intraoperative images for navigation correction.

[0044] To provide users with more path options, at least one neighboring candidate path can be generated in the vicinity of the initial needle insertion path. The target point of the neighboring candidate path is the same as that of the initial needle insertion path, but the insertion point is different. The insertion point of the neighboring candidate path serves as the candidate insertion point, and the insertion point of the initial needle insertion path serves as the initial insertion point. The candidate insertion points are located in the vicinity of the initial insertion point. Specifically, based on the initial needle insertion path, within a certain range, rays are generated at equal angular intervals around the initial needle insertion path, with the target point as the endpoint. The intersection of the rays with the skin is the candidate insertion point. Neighboring candidate paths are generated based on the target point and the candidate insertion points. Alternatively, neighboring candidate paths can be generated based on the initial needle insertion path and the medical image slice in which the initial needle insertion path is located. The target point of the initial needle insertion path serves as the target point of the neighboring candidate path. There is an angle between the neighboring candidate path and the initial needle insertion path, and the neighboring candidate path and the initial needle insertion path are in the same medical image slice.

[0045] In one embodiment, such as Figure 3As shown in process 300, processor 104 execution step 220 further includes performing the following steps:

[0046] Step 310: Obtain the plane that passes through the initial needle insertion point and is perpendicular to the initial needle insertion path.

[0047] Step 320: Based on the initial needle entry point and the preset range, determine the candidate region on the plane, generate discrete points on the candidate region, and generate candidate needle entry points based on the discrete points.

[0048] Step 330: Based on the target point and candidate needle entry points, generate at least one neighborhood candidate path.

[0049] Given the initial needle insertion path ET, where E is the initial needle insertion point and T is the target point.

[0050] a) Obtain PlaneA, which passes through point E and is perpendicular to the initial needle path ET.

[0051] (b) Using point E as the origin and a neighborhood search radius R as the preset range, candidate regions are generated on plane Plane A. Discrete points are selected within these candidate regions. The intersection of the line connecting the target point and the discrete point with the patient model's skin surface is used as candidate needle insertion points. If there is only one discrete point, one candidate needle insertion point is generated; if there are multiple discrete points, multiple candidate needle insertion points are generated. The preset range can be 2cm, or any value greater than 5mm and less than 3cm.

[0052] c) Connect the target point and a candidate needle entry point to generate a neighborhood candidate path, and connect the target point and multiple candidate needle entry points to generate multiple neighborhood candidate paths.

[0053] Discrete points can be generated synchronously on the candidate region; discrete points can also be generated by shrinking the radius R inward, i.e., there are multiple R values, which are generated step by step on multiple circles with point E as the origin and R as the radius.

[0054] Traditional methods only verify the feasibility of a single path, without providing overall executability information within the neighborhood. Users need to repeatedly try different paths to find a suitable one. This new method generates candidate entry points based on the initial entry point, and at least one neighborhood candidate path based on the target point and candidate entry points. The introduction of a neighborhood range overcomes the limitations of a single path, providing a foundation for searching for the target path among neighborhood candidate paths. This augmented simulation approach not only comprehensively evaluates path feasibility but also provides users with more adjustment references, thereby significantly reducing the time cost of path planning.

[0055] Since interventional surgery is performed with the assistance of a robotic arm, the joint angles of the robotic arm must be solvable, and obstacle avoidance by the robotic arm must be collision-free. The neighborhood candidate paths obtained in the above embodiments also need to be verified to satisfy the lightweight constraints of the robotic arm's joint angles and collision constraints.

[0056] Taking a five-joint robotic arm as an example, the pose of the robotic arm's end effector at the starting point can be described by the joint angle values ​​of its five joints, referred to as the initial joint angles of each joint. The pose of the robotic arm's end effector at the target point can also be described by the joint angle values ​​of its five joints, referred to as the target joint angles of each joint. Specifically, the initial joint angles of each joint refer to the joint angles of the robotic arm when the end effector reaches the starting point; this is a set of joint angles, which can be represented by O0[q1,q2,q3,q4,q5]. The target joint angles of each joint refer to the joint angles of the robotic arm when the end effector reaches the target point; this is a set of joint angles, referred to by Q0[q1,q2,q3,q4,q5]. m [q1,q2,q3,q4,q5] represents this.

[0057] The processor can use a preset robotic arm path planning algorithm to plan the path for each joint angle of the robotic arm to transform from the starting joint angle to the target joint angle. This preset algorithm can be a joint linear interpolation algorithm, such as the A* algorithm, Dijkstra's algorithm, D* algorithm, or artificial potential field method. This algorithm can uniformly insert several intermediate joint angles between the starting joint angle at the origin and the target joint angle at the destination, with a certain step size. It can also be a sampling search algorithm, such as the Probabilistic Roadmap Method (PRM) or Rapidly-exploring Random Tree (RRT). Furthermore, it can be an intelligent acoustic path planning algorithm, such as a neural network algorithm, ant colony algorithm, or genetic algorithm, or other path planning algorithms. When the robotic arm can successfully transform from the starting joint angle to the target joint angle, regardless of whether the intermediate joint angles are reachable, this is said to satisfy the lightweight constraint of the robotic arm joint angles. When a robotic arm can successfully transform from the initial joint angle to the joint angle at the intermediate point, and further transform from the joint angle at the intermediate point to the target joint angle, it is said to satisfy the full constraint of the robotic arm joint angles. When there are multiple paths planned from the starting point to the target point, i.e., multiple candidate paths in the neighborhood, and the target point is a candidate needle entry point, it is possible to determine whether it satisfies the lightweight constraint of the robotic arm joint angles, without having to perform full constraint calculations, which can significantly reduce the amount of computation and improve computational efficiency.

[0058] Further collision detection is required for all joint angles, including the starting joint angle, intermediate joint angles, and target joint angle. If the robotic arm and its end effector corresponding to each joint angle do not collide with the environment, the path satisfies the collision constraints. Alternatively, the collision detection step for intermediate joint angles can be removed; it is only necessary to confirm that the robotic arm and its end effector corresponding to the target joint angle do not collide with the environment to be considered to satisfy the collision constraints. The environment can refer to the imaging equipment model or the patient model. When there are multiple paths planned from the starting point to the target point (i.e., multiple candidate paths in the neighborhood), and the target point is a candidate needle insertion point, it is only necessary to confirm that the robotic arm and its end effector corresponding to the target joint angle do not collide with the environment. Collision detection for intermediate joint angles is not required, which can significantly reduce the computational load and improve computational efficiency.

[0059] It should be understood that the movement of the robotic arm and its end effector, the joint angles and poses described by this system in order to find the target path are simulated. No actual movement, joint angle rotation and pose change occur in physical space.

[0060] In one embodiment, processor execution step 230 further includes performing the following steps:

[0061] Step 410: Construct a first pose error matrix based on the first pose, the second pose, and the first joint angle; solve for the second joint angle based on the first pose error matrix. The first pose is used to simulate the pose of the end effector of the robotic arm at the initial point of the iteration; the first joint angle is used to simulate the joint angle of the robotic arm when the end effector of the robotic arm is in the first pose; the second pose is used to simulate the pose of the end effector of the robotic arm when it is in the candidate needle entry point of the neighboring candidate path; and the second joint angle is used to simulate the joint angle of the robotic arm when the end effector of the robotic arm is in the second pose.

[0062] Step 420: Determine whether the candidate path in the neighborhood satisfies the lightweight constraint of the robot arm joint angle. If the lightweight constraint of the robot arm joint angle is satisfied, the second joint angle has a solution when the end of the robot arm is in the second pose. If the lightweight constraint of the robot arm joint angle is not satisfied, the second joint angle has no solution when the end of the robot arm is in the second pose.

[0063] Step 430: When there is a solution for the second joint angle, determine whether the collision constraint is satisfied when the end of the robotic arm is in the second pose.

[0064] Step 440: Select the neighborhood candidate paths that satisfy the lightweight constraints and collision constraints of the robot arm joint angle as the target paths, wherein there are one or more target paths.

[0065] The first pose can be any pose, such as the pose of the robotic arm during registration and alignment with the imaging device. The first joint angle is the joint angle of the robotic arm when its end effector is in the first pose. The second pose is the pose of the target point mentioned above, and the second joint angle is the target joint angle mentioned above. A first pose error matrix is ​​constructed based on the first pose, second pose, and first joint angle. Using the robotic arm inverse kinematics algorithm, the first pose error matrix is ​​iteratively updated to predict and update the first pose until it reaches zero. The second joint angle is then solved. If a solution exists for the second joint angle, the neighboring candidate path satisfies the lightweight constraint of the robotic arm joint angle; otherwise, the neighboring candidate path does not satisfy the lightweight constraint. It can be understood that the second joint angle of a multi-joint robotic arm contains multiple joint angles, and "no solution for the second joint angle" means that at least one of the second joint angles is unsolvable. For example, in a five-joint robotic arm, the second joint angle includes five joint angles.

[0066] Specifically, based on the inverse kinematics algorithm of the robotic arm, the inverse kinematics equation is established by combining the first pose T1, the first joint angle, and the second pose T2 of the robotic arm end effector, that is, the first pose error matrix is ​​constructed: T 12 This represents the pose change from T1 to T2, where T... 12 Including position component P 12 and rotational component R 12 According to the position component P 12 and rotational component R 12 Iterate T1, making T 12 =0, solve for the second joint angle Q2. The specific calculation uses numerical methods, such as the Newton-Raphson method iterative solution.

[0067] The solution for the second joint angle can be further constrained based on the robotic arm joints, ensuring that the solution for the second joint angle is within the range of rotation of the robotic arm joints. Here, Q2 can be used to refer to Q mentioned above. m .

[0068] When a solution for the second joint angle is available, it is necessary to further determine whether there is a collision risk between the robotic arm and the environment. If there are multiple candidate paths in the neighborhood and multiple second poses, to reduce computational load and improve efficiency, the collision risk during the movement of the robotic arm and its end effector from the first pose to the second pose can be ignored. Only the collision risk between the robotic arm and its end effector and the environment when the end effector is in the second pose is calculated. Satisfying the collision constraint means that when the end effector reaches the second pose, the robotic arm and its end effector do not interfere with the imaging device model, and the robotic arm and its end effector do not interfere with the patient model. Of course, when computational resources are sufficient, such as when there is only one or a few candidate paths in the neighborhood, the collision risk during the movement of the robotic arm and its end effector from the first pose to the second pose can be considered. Satisfying the collision constraint means that during the movement of the robotic arm from the first pose to the second pose, the robotic arm and its end effector do not interfere with the imaging device model, and the robotic arm and its end effector do not interfere with the patient model; and when the end effector reaches the second pose, the robotic arm and its end effector do not interfere with the imaging device model, and the robotic arm and its end effector do not interfere with the patient model.

[0069] Neighborhood candidate paths that satisfy the lightweight constraints and collision constraints of the robotic arm's joint angles are selected as target paths. There can be one or more target paths. Based on the above system, and by analyzing the actual working scenario of the robotic arm, the complex joint angle calculation and collision model are reasonably simplified. Lightweight constraints and collision constraints of the robotic arm's joint angles are adopted, retaining key parameters while reducing computational complexity.

[0070] The candidate needle entry point is within the neighborhood of the initial needle entry point. The joint angles of the robotic arm at the initial needle entry point are similar to those at the candidate needle entry points. To reduce computational load and improve efficiency, the joint angles of the robotic arm have a solution when the end of the robotic arm along the initial needle entry path is at the initial needle entry point. When calculating the second joint angle of the robotic arm, the joint angle at the initial needle entry point can be used to quickly calculate the second joint angle using a fast inverse kinematics method. The specific method is as follows.

[0071] In one embodiment, the processor further includes performing the following steps:

[0072] Step 510: Construct a second pose error matrix based on the first pose, the third pose, and the first joint angle; solve for the third joint angle based on the second pose error matrix; the third pose is used to simulate the pose of the robotic arm end effector when it is located at the initial needle insertion point of the initial needle insertion path; the third joint angle is used to simulate the joint angle of the robotic arm when the robotic arm end effector is located in the third pose.

[0073] Step 520: When the end effector of the robotic arm is in the third pose, the third joint angle has a solution. The third joint angle is used as the initial solution and introduced into the first pose error matrix to calculate the second joint angle.

[0074] Specifically, the first pose is used to simulate the pose of the robotic arm's end effector at the initial point of the iteration. The first pose can be any pose, such as the pose of the robotic arm during registration with the imaging device. Based on the robotic arm inverse kinematics algorithm, the inverse kinematics equations are established by combining the first pose T1 and the third pose T3 of the robotic arm's end effector, thus constructing the second pose error matrix: T 13 This represents the pose change from T1 to T3, where T... 13 Including position component P 13 and rotational component R 13 According to the position component P 13 and rotational component R 13 Iterate T1, making T 13 =0, solve for the third joint angle Q3. The specific calculation uses numerical methods, such as the Newton-Raphson method iterative solution.

[0075] The solution for the third joint angle can be further constrained based on the joints of the robotic arm, so that the solution for the third joint angle is within the range of rotation of the robotic arm joint.

[0076] If a valid solution exists for Q3, then using numerical iteration or numerical optimization methods, O3 is introduced as an initial solution into the first pose error matrix to calculate the second joint angle. Introducing Q3 as an initial solution into the first pose error matrix can be understood as replacing T1 with T3 to solve for Q2, thereby optimizing the solution efficiency of Q2.

[0077] When there are multiple candidate paths in the neighborhood, there are also multiple candidate needle entry points for each candidate needle entry point, and there are also multiple poses corresponding to each candidate needle entry point. That is, there are multiple second poses for the robotic arm to reach the candidate needle entry point, and there are multiple sets of second joint angles of the robotic arm. In order to improve the calculation efficiency of the second joint angle, similar to the above, an initial solution is introduced to accelerate the solution efficiency.

[0078] In one embodiment, there are multiple candidate paths in the neighborhood, multiple candidate needle entry points for the candidate paths in the neighborhood, multiple second poses for the robotic arm end effector, and multiple first pose error matrices. The first pose error matrices correspond one-to-one with the candidate paths in the neighborhood. The processor is also used to perform the following steps:

[0079] Step 610: Iterate through and determine whether each neighborhood candidate path satisfies the lightweight constraint of the robot arm joint angle. When any neighborhood candidate path satisfies the lightweight constraint of the robot arm joint angle, the end of the robot arm is located in a sub-pose. The second pose includes the sub-pose. The sub-pose error matrix has a solution for the sub-joint angle. The first pose error matrix includes the sub-pose error matrix. The second joint angle includes the sub-joint angle.

[0080] Step 620: The sub-joint angle is used as the initial solution to calculate the corresponding second joint angle in other first-position error matrices. The other first-position error matrices are the first-position error matrices corresponding to the untraversed neighborhood candidate paths.

[0081] Specifically, if there are multiple candidate paths in the neighborhood, then there are multiple second poses, each including a sub-pose. There are multiple sets of second joint angles, each including a sub-joint angle. Multiple first pose error matrices are constructed from the first pose and the multiple second poses. Each first pose error matrix includes a sub-pose error matrix, which is constructed from the first pose and the sub-poses. The solution for the robot arm's joint angles under a sub-pose is the sub-joint angle. The first pose corresponding to each of the multiple first pose error matrices can be the same or different. The same first pose corresponds to the same first joint angle, and different first poses correspond to different joint angles. The first pose is used to simulate the pose of the robot arm's end effector at the initial point of iteration. The first pose can be any pose, such as the pose of the robot arm during registration and alignment with the imaging device.

[0082] We can iterate through each neighborhood candidate path to determine if it satisfies the lightweight constraints on the robot arm's joint angles. If a neighboring candidate path satisfies these constraints, the robot arm's end effector is in a sub-pose, and the solution for the joint angles is the sub-joint angle. This sub-joint angle is then used as the initial solution and introduced into the calculation of the lightweight constraints on the remaining untraversed neighboring candidate paths. Specifically, it incorporates the first pose error matrix corresponding to the untraversed neighboring candidate paths to optimize the efficiency of solving the remaining second joint angles. In essence, the sub-pose is used to replace the first pose in the calculation of the first pose error matrix for the untraversed paths, yielding the corresponding second joint angle.

[0083] Based on the above, neighborhood candidate paths that satisfy the lightweight constraints and collision constraints of the robotic arm joint angles are used as target paths. There can be one or more target paths. Intervention paths are determined based on these target paths, and the intervention paths characterize the planning results used for the interventional surgery. Specifically, the user can select one or more target paths as intervention paths; or the processor can directly recommend one or more intervention paths from the target paths. It can be understood that when the interventional surgery is a single-needle procedure, one of the target paths is used as the intervention path. When the interventional surgery is a multi-needle procedure, multiple target paths are used as intervention paths. However, when there is only one target path, it can be directly used as the intervention path.

[0084] The small intervals between multiple target paths mean that any path within the path region formed by these target paths can be used as an intervention path. Therefore, the intervention path is not limited to the target paths themselves. Thus, an intervention point can be selected within the target insertion area formed by the target insertion points of the target paths, thereby determining the intervention path.

[0085] In one embodiment, the processor is also configured to perform the following steps:

[0086] Step 710: Obtain the target needle entry point of the target path;

[0087] Step 720: Determine the target needle insertion area based on the target needle insertion point, select an intervention needle insertion point within the target needle insertion area, and determine the intervention path based on the intervention needle insertion point and the target point. There may be one or more intervention paths.

[0088] Specifically, the target needle insertion points within the target path form the target needle insertion area. Users can actively select one or more interventional needle insertion points within this area, and the interventional path is determined by the line connecting the interventional needle insertion points and the target point. The processor can also automatically select one or more interventional needle insertion points within the target needle insertion area, and determine the interventional path by the line connecting the interventional needle insertion points and the target point. In other words, when the interventional procedure is a single-needle procedure, one interventional needle insertion point can be selected, and one interventional path can be determined. When the interventional procedure is a multi-needle procedure, multiple interventional needle insertion points can be selected, and multiple interventional paths can be determined. In the above system, any path within the path area formed by multiple target paths can be used as an interventional path, thus expanding the range of selectable interventional paths.

[0089] Furthermore, based on the lightweight constraints of the robotic arm's joint angles and collision constraints, candidate paths in the neighborhood are filtered. Paths that satisfy the constraints are designated as target paths, while those that do not are designated as non-target paths. To clearly distinguish between the two and facilitate intuitive user identification, different colors can be used to differentiate between target and non-target paths. Specifically, target paths can be marked in green, and non-target paths in red. Of course, other colors can also be used, as long as they can distinguish between target and non-target paths.

[0090] The target path includes the target needle insertion point, and the non-target path includes non-target needle insertion points. Different colors can be used to distinguish the target needle insertion area (composed of target needle insertion points) from the non-target needle insertion area (composed of non-target needle insertion points), making it easy for users to quickly and intuitively identify the two areas. Furthermore, the intervention needle insertion point can be quickly determined on the target needle insertion area to confirm the intervention path.

[0091] In one embodiment, the processor is also configured to perform the following steps:

[0092] Different colors are used to represent the target needle entry area and the non-target needle entry area. The neighborhood candidate path consists of the target path and the non-target path. The non-target needle entry area consists of the needle entry points of the non-target path.

[0093] Specifically, the target needle insertion area can be marked in green, and the non-target needle insertion area in red. Of course, other colors can also be used, as long as they can distinguish between the target and non-target needle insertion areas.

[0094] Furthermore, if the initial needle insertion path changes, according to the description in the above system, the neighborhood candidate path is related to the initial needle insertion path. Therefore, if the neighborhood candidate path changes, the target path and non-target path generated based on the neighborhood candidate path will also change, and thus the target needle insertion area and non-target needle insertion area will change.

[0095] In one embodiment, the processor is also configured to perform the following steps:

[0096] Adjust the initial needle insertion path;

[0097] The target needle insertion area and the non-target needle insertion area change as the initial needle insertion path is adjusted.

[0098] The target path and non-target paths are still distinguished by different colors after further changes.

[0099] Similarly, the changed target needle insertion area and non-target needle insertion area are still distinguished by different colors.

[0100] In one embodiment, an interventional surgical path planning method is provided, which includes the following steps:

[0101] Obtain the initial needle insertion path for the interventional procedure, which includes the target point and the initial needle insertion point;

[0102] Candidate needle insertion points are generated based on the initial needle insertion point;

[0103] Based on the target point and candidate needle entry points, at least one neighborhood candidate path is generated.

[0104] The target path is determined based on neighborhood candidate paths, lightweight constraints on the joint angles of the robotic arm, and collision constraints.

[0105] In one embodiment, the method of generating candidate needle entry points based on an initial needle entry point and generating at least one neighborhood candidate path based on the target point and candidate needle entry points further includes the following steps:

[0106] Obtain a plane that passes through the initial needle insertion point and is perpendicular to the initial needle insertion path;

[0107] Based on the initial needle insertion point and the preset range, a candidate region is determined on the plane, discrete points are generated on the candidate region, and the candidate needle insertion point is generated based on the discrete points.

[0108] Based on the target point and candidate needle entry points, at least one neighborhood candidate path is generated.

[0109] In one embodiment, the method for determining the target path based on neighborhood candidate paths, lightweight constraints on robot joint angles, and collision constraints further includes the following steps:

[0110] The first pose error matrix is ​​constructed based on the first pose, the second pose, and the first joint angle. The second joint angle is solved based on the first pose error matrix. The first pose is used to simulate the pose of the end effector of the robot arm when it is located at the initial point of the iteration. The first joint angle is used to simulate the joint angle of the robot arm when it is located at the first pose. The second pose is used to simulate the pose of the end effector of the robot arm when it is located at the candidate needle entry point of the candidate path in the neighborhood. The second joint angle is used to simulate the joint angle of the robot arm when it is located at the second pose.

[0111] Determine whether the candidate paths in the neighborhood satisfy the lightweight constraint of the robot arm joint angle. If the lightweight constraint of the robot arm joint angle is satisfied, the second joint angle has a solution when the end of the robot arm is in the second pose. If the lightweight constraint of the robot arm joint angle is not satisfied, the second joint angle has no solution when the end of the robot arm is in the second pose.

[0112] When there is a solution for the second joint angle, determine whether the collision constraint is satisfied when the end of the robotic arm is in the second pose.

[0113] The neighborhood candidate paths that satisfy the lightweight constraints of the robot arm's joint angles and collision constraints are used as the target paths, and there can be one or more target paths.

[0114] In one embodiment, the method further includes the following steps:

[0115] A second pose error matrix is ​​constructed based on the first pose, the third pose, and the first joint angle. The third joint angle is solved based on the second pose error matrix. The third pose is used to simulate the pose of the end effector of the robotic arm when it is located at the initial needle insertion point of the initial needle insertion path. The third joint angle is used to simulate the joint angle of the robotic arm when the end effector of the robotic arm is located in the third pose.

[0116] When the end effector of the robotic arm is in the third pose, the third joint angle has a solution. The third joint angle is used as the initial solution and introduced into the first pose error matrix to calculate the second joint angle.

[0117] In one embodiment, there are multiple candidate paths in the neighborhood, multiple candidate needle entry points for the candidate paths in the neighborhood, multiple second poses of the robotic arm end effector, and multiple first pose error matrices. The first pose error matrix corresponds one-to-one with the candidate paths in the neighborhood. The method further includes the following steps:

[0118] Iterate through each neighborhood candidate path to determine whether it satisfies the lightweight constraint of the robot arm joint angle. When any neighborhood candidate path satisfies the lightweight constraint of the robot arm joint angle, the end of the robot arm is in a sub-pose. The second pose includes the sub-pose. The sub-pose error matrix has a solution for the sub-joint angle. The first pose error matrix includes the sub-pose error matrix. The second joint angle includes the sub-joint angle.

[0119] The sub-joint angle is used as the initial solution to calculate the corresponding second joint angle in other first-position error matrices. The other first-position error matrices are the first-position error matrices corresponding to the untraversed neighborhood candidate paths.

[0120] In one embodiment, the method further includes the following steps:

[0121] Obtain the target infeed point of the target path;

[0122] The target needle insertion area is determined based on the target needle insertion point. An intervention needle insertion point is selected within the target needle insertion area. The intervention path is determined based on the intervention needle insertion point and the target point. There can be one or more intervention paths.

[0123] In one embodiment, the method further includes the following steps:

[0124] Different colors are used to represent the target needle entry area and the non-target needle entry area. The neighborhood candidate path consists of the target path and the non-target path. The non-target needle entry area consists of the needle entry points of the non-target path.

[0125] In one embodiment, the method further includes the following steps:

[0126] Adjust the initial needle insertion path;

[0127] The target needle insertion area and the non-target needle insertion area change as the initial needle insertion path is adjusted.

[0128] It should be understood that, although Figure 2-3 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 2-3 At least some of the steps in the process may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but may be executed at different times. The execution order of these steps or stages is not necessarily sequential, but may be executed in turn or alternately with other steps or at least some of the steps or stages in other steps.

[0129] In one embodiment, such as Figure 4As shown, an interventional surgical path planning device 800 is provided. This device includes an initial needle insertion path acquisition module 810, a candidate needle insertion point generation module 820, a neighborhood candidate path generation module 830, and a target path determination module 840, wherein:

[0130] The initial needle insertion path acquisition module 810 is used to acquire the initial needle insertion path of the interventional procedure. The initial needle insertion path includes the target point and the initial needle insertion point.

[0131] The candidate needle insertion point generation module 820 generates candidate needle insertion points based on the initial needle insertion point;

[0132] The neighborhood candidate path generation module 830 generates at least one neighborhood candidate path based on the target point and candidate needle entry points.

[0133] The target path determination module 840 determines the target path based on neighborhood candidate paths, lightweight constraints on the joint angles of the robotic arm, and collision constraints.

[0134] Specific limitations regarding the interventional surgical pathway planning device can be found in the above description of the system's limitations, and will not be repeated here. Each module in the aforementioned interventional surgical pathway planning device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the computer device's memory as software, so that the processor can call and execute the corresponding operations of each module.

[0135] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 5 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements an interventional surgical path planning method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0136] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0137] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0138] Obtain the initial needle insertion path for the interventional procedure, which includes the target point and the initial needle insertion point;

[0139] Candidate needle insertion points are generated based on the initial needle insertion point;

[0140] Based on the target point and candidate needle entry points, at least one neighborhood candidate path is generated.

[0141] The target path is determined based on neighborhood candidate paths, lightweight constraints on the joint angles of the robotic arm, and collision constraints.

[0142] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0143] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0144] Obtain the initial needle insertion path for the interventional procedure, which includes the target point and the initial needle insertion point;

[0145] Candidate needle insertion points are generated based on the initial needle insertion point;

[0146] Based on the target point and candidate needle entry points, at least one neighborhood candidate path is generated.

[0147] The target path is determined based on neighborhood candidate paths, lightweight constraints on the joint angles of the robotic arm, and collision constraints.

[0148] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0149] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0150] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0151] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. An interventional surgical path planning system, characterized in that, The system includes a processor, the processor being configured to perform the following steps: Obtain the initial needle insertion path for interventional surgery, wherein the initial needle insertion path includes the target point and the initial needle insertion point; Based on the initial needle insertion point, candidate needle insertion points are generated, and based on the target point and the candidate needle insertion points, at least one neighborhood candidate path is generated. The target path is determined based on the neighborhood candidate paths, the lightweight constraints of the robot arm joint angles, and the collision constraints.

2. The system according to claim 1, characterized in that, The processor's execution of generating candidate needle entry points based on the initial needle entry point, and generating at least one neighborhood candidate path based on the target point and the candidate needle entry points, further includes performing the following steps: Obtain a plane that passes through the initial needle insertion point and is perpendicular to the initial needle insertion path; Based on the initial needle insertion point and the preset range, a candidate region is determined on the plane, discrete points are generated on the candidate region, and the candidate needle insertion point is generated based on the discrete points. Based on the target point and the candidate needle insertion point, at least one neighborhood candidate path is generated.

3. The system according to claim 1, characterized in that, The processor's determination of the target path based on the neighborhood candidate paths, lightweight constraints on the robot arm joint angles, and collision constraints further includes performing the following steps: A first pose error matrix is ​​constructed based on the first pose, the second pose, and the first joint angle. The second joint angle is solved based on the first pose error matrix. The first pose is used to simulate the pose of the end effector of the robotic arm at the initial point of the iteration. The first joint angle is used to simulate the joint angle of the robotic arm when the end effector of the robotic arm is in the first pose. The second pose is used to simulate the pose of the end effector of the robotic arm when it is in the candidate needle entry point of the neighborhood candidate path. The second joint angle is used to simulate the joint angle of the robotic arm when the end effector of the robotic arm is in the second pose. Determine whether the neighborhood candidate path satisfies the lightweight constraint of the robot arm joint angle. If the lightweight constraint of the robot arm joint angle is satisfied, the second joint angle has a solution when the end of the robot arm is in the second pose. If the lightweight constraint of the robot arm joint angle is not satisfied, the second joint angle has no solution when the end of the robot arm is in the second pose. When the second joint angle has a solution, determine whether the collision constraint is satisfied when the end of the robotic arm is in the second pose; The neighborhood candidate paths that satisfy the lightweight constraints of the robotic arm joint angles and the collision constraints are taken as the target paths, and there are one or more target paths.

4. The system according to claim 3, characterized in that, The processor is also used to perform the following steps: A second pose error matrix is ​​constructed based on the first pose, the third pose, and the first joint angle. The third joint angle is solved based on the second pose error matrix. The third pose is used to simulate the pose of the end effector of the robotic arm when it is located at the initial needle insertion point of the initial needle insertion path. The third joint angle is used to simulate the joint angle of the robotic arm when the end effector of the robotic arm is located in the third pose. When the end effector of the robotic arm is in the third pose, the third joint angle has a solution. The third joint angle is used as the initial solution and introduced into the first pose error matrix to calculate the second joint angle.

5. The system according to claim 3, characterized in that, There are multiple candidate paths in the neighborhood, multiple candidate needle entry points in the neighborhood candidate paths, multiple second poses of the robotic arm end effector, and multiple first pose error matrices. The first pose error matrices correspond one-to-one with the neighborhood candidate paths. The processor is further configured to perform the following steps: Iterate through each of the neighborhood candidate paths to determine whether it satisfies the lightweight constraint of the robot arm joint angle. When any of the neighborhood candidate paths satisfies the lightweight constraint of the robot arm joint angle, the end of the robot arm is located in a sub-pose. The second pose includes the sub-pose. The sub-joint angle of the sub-pose error matrix has a solution. The first pose error matrix includes the sub-pose error matrix. The second joint angle includes the sub-joint angle. The subjoint angle is used as the initial solution to calculate the corresponding second joint angle by introducing other first-order pose error matrices. The other first-order pose error matrices are the first-order pose error matrices corresponding to the untraversed neighborhood candidate paths.

6. The system according to claim 1, characterized in that, The processor is also used to perform the following steps: Obtain the target needle entry point of the target path; The target needle insertion area is determined based on the target needle insertion point. An intervention needle insertion point is selected within the target needle insertion area. An intervention path is determined based on the intervention needle insertion point and the target point. The intervention path may be one or more.

7. The system according to claim 6, characterized in that, The processor is also used to perform the following steps: The target needle insertion area and the non-target needle insertion area are represented by different colors. The neighborhood candidate path is composed of the target path and the non-target path. The non-target needle insertion area is composed of the needle insertion points of the non-target path.

8. The system according to claim 7, characterized in that, The processor is also used to perform the following steps: Adjust the initial needle insertion path; The target needle insertion area and the non-target needle insertion area change as the initial needle insertion path is adjusted.

9. A method for planning interventional surgical pathways, characterized in that, The method includes: Obtain the initial needle insertion path for interventional surgery, wherein the initial needle insertion path includes the target point and the initial needle insertion point; Candidate needle insertion points are generated based on the initial needle insertion point; Based on the target point and the candidate needle insertion point, at least one neighborhood candidate path is generated; The target path is determined based on the neighborhood candidate paths, the lightweight constraints of the robot arm joint angles, and the collision constraints.

10. An interventional surgical path planning device, characterized in that, The device includes: The initial needle insertion path acquisition module is used to acquire the initial needle insertion path of the interventional procedure, wherein the initial needle insertion path includes the target point and the initial needle insertion point; The candidate needle insertion point generation module generates candidate needle insertion points based on the initial needle insertion point; The neighborhood candidate path generation module generates at least one neighborhood candidate path based on the target point and the candidate needle entry point. The target path determination module determines the target path based on the neighborhood candidate paths, the lightweight constraints of the robot arm joint angles, and the collision constraints.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method of claim 9.

Images