CN-122005095-A - Surgical robot autonomous intervention instrument navigation method, device and system

CN122005095ACN 122005095 ACN122005095 ACN 122005095ACN-122005095-A

Abstract

The invention discloses a surgical robot autonomous intervention instrument navigation method, a device and a system, and relates to the field of surgical robots, wherein the method comprises the steps of S1 obtaining preoperative CT data, S2 extracting a three-dimensional geometric model and a topological structure of a vascular tree to generate a path containing a central line and a lumen size, S3 collecting three-dimensional space position and gesture of an intervention instrument and an intra-operative ultrasonic image in real time, S4 aligning the intra-operative ultrasonic image, the three-dimensional space position and gesture and the three-dimensional geometric model and mapping the three-dimensional space position and gesture and the three-dimensional geometric model to a unified coordinate system, S5 constructing a final navigation path, S6 predicting, controlling and guiding wire delivery, and constructing a 'perception-guidance-automatic path-searching-execution-monitoring' full-closed loop autonomous intervention operation system based on multi-modal information fusion.

Inventors

GU LIXU
MA YUXIANG
ZHANG YINGJIE
ZHANG FAQIANG

Assignees

华西精创医疗科技(成都)有限公司

Dates

Publication Date: 20260512
Application Date: 20260410

Claims (9)

1. The autonomous intervention instrument navigation method of the surgical robot is characterized by comprising the following steps: s1, acquiring preoperative CT data of a person to be operated; s2, extracting a three-dimensional geometric model and a topological structure of a blood vessel tree according to preoperative CT data, and generating a model comprising a central line path and a lumen size; s3, acquiring the three-dimensional space position and the three-dimensional space gesture of the interventional instrument in real time, and acquiring an intraoperative ultrasonic image; S4, aligning the three-dimensional space position and posture of the intraoperative ultrasonic image and the interventional instrument with the three-dimensional geometric model, and mapping the three-dimensional space position and posture and the three-dimensional geometric model to a unified coordinate system; S5, constructing a final navigation path according to the central line path ; And S6, controlling the MPC to control the delivery of the guide wire according to the final navigation path by adopting model prediction.
2. The surgical robot-based autonomous interventional instrument navigation method according to claim 1, comprising in S5: S501, taking a blood vessel entrance node S as a starting point, taking a target node T of a surgical blood vessel as an end point, and constructing an initial navigation Path sequence as a Path, wherein the initial navigation Path sequence is expressed as: initial path node The iteration number m=0; S502, judging whether the distance between the current path node P m and the target node T is smaller than a distance threshold, if so, taking the current path sequence as a final navigation path And S6, otherwise, S503 is carried out; S503, obtaining three-dimensional state vectors of all candidate nodes of the next path node P m+1 according to the current path node P m ; S504, screening all candidate nodes by using constraint conditions to obtain a screened candidate node set; s505, according to the three-dimensional state vector of the candidate node Making a path decision to generate a local path between the current path node P m and the next path node P m+1 ; S506, updating the path sequence according to the local path to obtain an updated path sequence; S507, let m=m+1, and return to S502.
3. The surgical robot-based autonomous interventional instrument navigation method according to claim 2, wherein the three-dimensional state vector is included in S503 Expressed as: Wherein L (n) is in a geometrically reachable state, the geometrically reachable state is reachable or unreachable, when the vessel diameter meets the instrument passing condition and the space direction is reachable, the L (n) is reachable, otherwise the L (n) is unreachable, R (n) is in an environment risk state, the environment risk state is Safe, solution or Forbidden, when the vessel structure is stable, the R (n) is Safe, when plaque or abnormal vessel structure exists, the R (n) is Caution, when severe stenosis or high risk area exists, the R (n) is Forbidden, the M (n) is in a motion matching state, the motion matching state is stable or deviated, and when the guide wire motion direction vector is used Vector with the centerline of blood vessel Included angle of (2) Whether or not it is smaller than a preset threshold value If so, the value of M (n) is stable, otherwise, the value of M (n) is deviated and the included angle is formed Expressed as: 。
4. The method for navigating an autonomous surgical robot intervention instrument as claimed in claim 3, wherein in S504, the constraint conditions include condition 1, condition 2 and condition 3, wherein the constraint condition 1 is L (n) is unreachable, the constraint condition 2 is R (n) is Forbidden, and the constraint condition 3 is R (n) is Caution, wherein the filtering of the candidate nodes is specifically performed when the three-dimensional state vector of the candidate node is Deleting the candidate node when the condition 1 is satisfied, and not performing path expansion, wherein when the three-dimensional state vector of the candidate node is Deleting the candidate node when the condition 2 is satisfied, preventing the path from extending to the area, and not expanding the path, and when the three-dimensional state vector of the candidate node When the condition 3 is met, checking the number of continuous Caution-level risk nodes accumulated in the current path, adding the candidate nodes into a candidate node set if the accumulated number is smaller than a first preset number, deleting the candidate nodes if the accumulated number is equal to the first preset number, prohibiting the path from expanding to a continuous excessive risk area, and when the three-dimensional state vector of the candidate nodes is equal to the three-dimensional state vector of the candidate nodes And for the conditions 1 and 2, when the value of R (n) is Safe, resetting the number of the continuous risk nodes accumulated by the current path to a second preset number (recommended to be 0), and adding the candidate nodes into the candidate node set R, wherein the first preset number is smaller than the second preset number.
5. A surgical robot autonomous interventional instrument navigation method according to claim 3, characterized by comprising in S505: 1) Judging whether M (n) values of candidate nodes are stable, if so, keeping the current path direction continuously expanding, and selecting the candidate node with the smallest included angle with the current motion direction as the next path node And generate the current path node With the next path node Local paths between the two paths, and then entering S506, otherwise entering 2); 2) Judging whether the current adjustment times reach the preset times, if so, entering 4), otherwise, entering 3), 3) Correcting the direction of the local path, and enabling the current adjustment times to be +1 and returning to 1); 4) And (3) carrying out local path re-planning, and enabling the current adjustment times to be 0 and returning to 1).
6. The method for guiding an autonomous interventional instrument of a surgical robot according to claim 1, wherein the local path re-planning in 4) is specifically performed by taking a safety node where a current guide wire is located as a starting point, taking a nearest safety node downstream of an original path as an ending point as a local re-planning interval, refreshing three-dimensional state vectors of all nodes in the interval All candidate nodes are screened using constraints and return to 1).
7. The surgical robot autonomous interventional instrument navigation method according to claim 1, wherein in S6 specifically comprises: S601, predicting the motion state of the guide wire according to the three-dimensional space position and the gesture of the interventional instrument , wherein, The six-degree-of-freedom pose of the guide wire tip at the kth control period moment; controlling the input quantity for the robot in the kth control period; In order to predict the obtained k+1 control period guide wire tip pose, A guide wire motion state prediction model; s602, final navigation path Mapping to continuous target pose sequence under robot control coordinate system As a reference trajectory for the MPC controller, wherein, The target pose corresponding to the guide wire at the path node P i ; s603, the MPC controller solves the optimization objective function through rolling optimization in each control period Obtaining optimal control sequence and optimizing objective function Expressed as: Where N is the predicted time domain length, The predicted guide wire pose at the k+i time, For the target pose of the reference path at time k + i, The variable quantity of the input is controlled for the (k+i) th control period, Q is the weight matrix of the pose tracking deviation term, A weight matrix for control input; S604, the MPC controller outputs the control input of the current period to drive the robot to execute.
8. Surgical robot is from intervention apparatus navigation device, characterized by comprising: The memory is used for storing computer programs; An actuator for executing a computer program stored in a memory, which computer program, when executed, implements the surgical robot autonomous interventional instrument navigation method according to any of claims 1-7.
9. Surgical robot autonomous intervention instrument navigation system for implementing a surgical robot autonomous intervention instrument navigation method according to any of the claims 1-7, comprising: The acquisition module is used for acquiring preoperative CT data of a person to be operated; the extraction module is used for extracting a three-dimensional geometric model and a topological structure of the blood vessel tree according to the preoperative CT data and generating a preoperative navigation basic data set containing blood vessel central line paths and lumen size information; an acquisition module; the three-dimensional space position and the three-dimensional space gesture of the interventional instrument are acquired in real time through the magnetic positioning system, and simultaneously, an intraoperative ultrasonic image is acquired in real time through an intravascular ultrasonic catheter; The alignment module is used for aligning the three-dimensional space position and posture of the ultrasonic image and the interventional instrument in operation with the three-dimensional geometric model and mapping the three-dimensional space position and posture and the three-dimensional geometric model to a unified coordinate system; a building module for building final navigation path according to the central line path ; The model predictive control MPC controls the delivery of the guide wire according to the final navigation path; The monitoring and early warning module is used for monitoring and early warning based on the safety distance between the interventional instrument and the blood vessel and a preset distance threshold value according to the three-dimensional space position and the gesture of the interventional instrument and the intra-operation ultrasonic image, and generating early warning information; the system comprises a human-computer interaction module, wherein the human-computer interaction module comprises a visual interface and an input module, the visual interface is used for displaying various information in the operation process, and the input module is used for inputting operation information of a surgeon.

Description

Surgical robot autonomous intervention instrument navigation method, device and system Technical Field The invention relates to the field of surgical robots, in particular to a method, a device and a system for navigating autonomous interventional instruments of a surgical robot. Background In recent years, vascular interventional robotics have developed rapidly, but their degree of autonomy still faces two major bottlenecks. Firstly, the existing interventional operation robots mostly adopt a teleoperation mode, and the core function of the existing interventional operation robots is to improve the stability and the accuracy of operation, and the autonomy of navigation and execution is limited. Secondly, although a navigation system based on a single-mode image (such as X-ray radiography) can realize preoperative or intraoperative path planning, planning and execution links are separated, a preset path can only be passively executed, an autonomous closed loop of sensing-guiding-automatic path searching-executing-monitoring cannot be formed by depending on real-time multi-dimensional information, accurate automatic path searching capability under dynamic guiding is lacking, and vascular environment change in the operation is difficult to adapt. Disclosure of Invention The invention aims to solve the problems and designs a navigation method, a navigation device and a navigation system for an autonomous intervention instrument of a surgical robot. The invention realizes the above purpose through the following technical scheme: A surgical robot autonomous interventional instrument navigation method, comprising: s1, acquiring preoperative CT data of a person to be operated; s2, extracting a three-dimensional geometric model and a topological structure of a blood vessel tree according to preoperative CT data, and generating a model comprising a central line path and a lumen size; s3, acquiring the three-dimensional space position and posture of the interventional instrument in real time and an intraoperative ultrasonic image; S4, aligning the three-dimensional space position and posture of the intraoperative ultrasonic image and the interventional instrument with the three-dimensional geometric model, and mapping the three-dimensional space position and posture and the three-dimensional geometric model to a unified coordinate system; S5, constructing a final navigation path according to the central line path ; And S6, controlling the MPC to control the delivery of the guide wire according to the final navigation path by adopting model prediction. A surgical robot autonomous interventional instrument navigation device, comprising: The memory is used for storing computer programs; the actuator is used for executing the computer program stored in the storage, and the navigation method of the surgical robot autonomous interventional instrument is realized when the computer program is executed. The surgical robot autonomous intervention instrument navigation system is used for realizing the surgical robot autonomous intervention instrument navigation method, and comprises the following steps: The acquisition module is used for acquiring preoperative CT data of a person to be operated; the extraction module is used for extracting a three-dimensional geometric model and a topological structure of the blood vessel tree according to preoperative CT data and generating a model comprising a central line path and a lumen size; the acquisition module is used for acquiring the three-dimensional space position and posture of the interventional instrument and the intraoperative ultrasonic image in real time; The alignment module is used for aligning the three-dimensional space position and posture of the ultrasonic image and the interventional instrument in operation with the three-dimensional geometric model and mapping the three-dimensional space position and posture and the three-dimensional geometric model to a unified coordinate system; a building module for building final navigation path according to the central line path ; The model predictive control MPC controls the delivery of the guide wire according to the final navigation path; The monitoring and early warning module is used for monitoring and early warning based on the safety distance between the interventional instrument and the blood vessel and a preset distance threshold value according to the three-dimensional space position and the gesture of the interventional instrument and the intra-operation ultrasonic image, and generating early warning information; the system comprises a human-computer interaction module, wherein the human-computer interaction module comprises a visual interface and an input module, the visual interface is used for displaying various information in the operation process, and the input module is used for inputting operation information of a surgeon. The method has the beneficial effects that a full-closed loop autonomous intervention ope