Percutaneous spine endoscopic surgery robot path planning method and system

Abstract

The invention provides a percutaneous spine endoscopic surgery robot path planning method and system, and the method comprises the steps: determining the pixel offset characteristics of a percutaneous spine after the tissue structure deformation in a current spine surgery according to deformation data in tissue structure information; determining each obstacle area generated by tissue deformation in the percutaneous spine according to the pixel offset characteristics and the current motion state of the surgical robot, and predicting the obstacle influence degree of the tissue deformation in the spinal operation at the current operation moment according to the distribution characteristics of each obstacle area; feedback constraint is carried out on the obstacle influence degree according to the execution deviation of the actual operation path, and confidence constraint of an obstacle avoidance strategy in the spinal operation is obtained; and performing dynamic feedback on the obstacle avoidance strategy of the surgical robot in the subsequent surgical path based on confidence constraint to obtain an obstacle avoidance path of the surgical robot in the spinal surgery. Based on the scheme, the dynamic feedback adjustment of the obstacle avoidance path in the surgical robot can be realized, so that the safety of the percutaneous spine endoscopic surgery can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of surgical path planning, and more specifically, to a percutaneous spinal endoscopic surgical robot path planning method and system. Background Art

[0002] Traditional spinal surgery requires a larger incision, which may lead to slow postoperative recovery, high risk of complications and greater pain for the patient. Percutaneous spinal endoscopy is a minimally invasive surgical method that introduces endoscopic equipment through a percutaneous route and starts the surgical operation with a smaller incision. It can complete the diagnosis and treatment of spinal lesions with minimal trauma. It combines the characteristics of endoscope magnified field of view and delicate operation. It is suitable for the treatment of diseases such as intervertebral disc herniation and spinal stenosis, and significantly reduces the patient's trauma and postoperative recovery time.

[0003] Existing static path planning is difficult to respond in real time to factors such as anatomical changes, bleeding and fluid accumulation, changes in surrounding tissue pressure, and individual differences among patients during spinal surgery. The real-time changes in anatomical structure and tissue deformation caused by fluid accumulation make it impossible for the static path planning method of the surgical robot to accurately adapt to the dynamic environment. In addition, individual differences among patients also make it difficult for the surgical robot to provide accurate personalized path planning, which in turn affects the safety and accuracy of the operation. By introducing a dynamic feedback adjustment mechanism, the surgical robot can sense and adjust the path in real time, effectively adapting to the complex changes in spinal surgery, thereby ensuring the safety of spinal surgery. Therefore, how to achieve dynamic feedback adjustment of the obstacle avoidance path in the surgical robot to improve the safety of percutaneous endoscopic spinal surgery is a difficult problem faced by the industry. Summary of the invention

[0004] The present application provides a percutaneous spinal endoscopic surgery robot path planning method and system, which can realize dynamic feedback adjustment of the obstacle avoidance path in the surgical robot, thereby improving the safety of percutaneous spinal endoscopic surgery.

[0005] In a first aspect, the present application provides a path planning method for a percutaneous spinal endoscopic surgery robot, the method comprising: Generate the initial surgical path of percutaneous spinal endoscopy through anatomical images of the spine, and monitor the tissue structure information of the percutaneous spine during spinal surgery when the surgical robot performs spinal surgery according to the initial surgical path; Determine pixel offset features of the percutaneous spine after the tissue structure is deformed in the current spinal surgery according to the deformation data in the tissue structure information, determine various obstacle areas caused by the tissue deformation in the percutaneous spine through the pixel offset features and the current motion state of the surgical robot, and then predict the obstacle influence degree of the tissue deformation in the spinal surgery at the current surgery moment according to the distribution characteristics of each obstacle area; Acquire the actual surgical path of the surgical robot in the current spinal surgery, perform feedback constraints on the obstacle influence according to the execution deviation between the actual surgical path and the initial surgical path, and obtain the confidence constraints of the obstacle avoidance strategy in the spinal surgery; Based on the confidence constraint of the obstacle avoidance strategy, the obstacle avoidance strategy of the surgical robot in the subsequent surgical path is dynamically fed back to obtain the obstacle avoidance path of the surgical robot in spinal surgery.

[0006] In some embodiments, determining the pixel shift characteristics of the percutaneous spine after the tissue structure is deformed in the current spinal surgery according to the deformation data in the tissue structure information specifically includes: Comparing the anatomical image of the spine with the tissue structure information to obtain a deformation field between the real-time image and the original image during spinal surgery; Dividing the deformation field into a plurality of local interest regions based on the regional range of the initial surgical path, and then determining deformation sub-features of tissue structures in each local interest region; The pixel offset features of the percutaneous spine after the tissue structure is deformed in the current spinal surgery are determined according to all the deformation sub-features.

[0007] In some embodiments, the various obstacle areas caused by tissue deformation in the percutaneous spine are determined by using the pixel offset feature and the current motion state of the surgical robot, specifically including: Obtain the current motion state of the surgical robot, and then predict the actual motion path of the surgical robot in the current motion state; Predicting the offset impact area in the actual motion path by using the pixel offset feature; dividing the deviation-affected region into a plurality of obstacle regions based on tissue characteristics in the percutaneous spine; Individual obstruction areas resulting from tissue deformation in the percutaneous spine are determined from all obstruction areas.

[0008] In some embodiments, predicting the obstacle influence of tissue deformation in spinal surgery at the current surgery time based on the distribution characteristics of each obstacle area specifically includes: Extract features from each obstacle area to obtain distribution features of each obstacle area; Determine the interference information of tissue deformation on spinal surgery through all distribution characteristics; Based on the obstacle interference information, a cross prediction is performed on the path interference in the initial surgical path to obtain the obstacle influence degree of tissue deformation in spinal surgery at the current surgical moment.

[0009] In some embodiments, feedback constraints are applied to the obstacle influence degree according to the execution deviation between the actual surgical path and the initial surgical path to obtain the confidence constraints of the obstacle avoidance strategy in spinal surgery, specifically including: Determining the execution deviation between the actual surgical path and the initial surgical path by a path comparison algorithm; Determining a feedback adjustment amount of the execution deviation on an actual surgical path; Determining the obstacle avoidance constraint amount of the obstacle influence on the actual surgical path; The confidence constraint of the obstacle avoidance strategy in spinal surgery is determined based on the feedback adjustment amount and the deviation constraint amount.

[0010] In some embodiments, the obstacle avoidance strategy of the surgical robot in the subsequent surgical path is dynamically fed back based on the confidence constraint of the obstacle avoidance strategy, and the obstacle avoidance path of the surgical robot in spinal surgery is obtained, which specifically includes: Initialize a dynamic obstacle avoidance model for a path planning algorithm; Determining the dynamic constraint conditions of the dynamic obstacle avoidance model through the confidence constraints of the obstacle avoidance strategy; Determine the target optimization function of the dynamic obstacle avoidance model through the obstacle avoidance strategy of the initial surgical path; The dynamic obstacle avoidance model is used to feedback-adjust the obstacle avoidance strategy of the surgical robot in the subsequent surgical path, and the obstacle avoidance path of the surgical robot in spinal surgery is obtained.

[0011] In some embodiments, ultrasound imaging equipment is used to monitor anatomic information of the spine during percutaneous spinal surgery.

[0012] In a second aspect, the present application provides a percutaneous spinal endoscopic surgery robot path planning system, comprising: A monitoring module, used to generate an initial surgical path for percutaneous spinal endoscopy through anatomical images of the spine, and to monitor the tissue structure information of the percutaneous spine during the spinal surgery when the surgical robot performs the spinal surgery according to the initial surgical path; A processing module, for determining pixel offset features of the percutaneous spine after the tissue structure is deformed in the current spinal surgery according to the deformation data in the tissue structure information, determining various obstacle areas caused by the tissue deformation in the percutaneous spine through the pixel offset features and the current motion state of the surgical robot, and then predicting the obstacle influence degree of the tissue deformation in the spinal surgery at the current surgery moment according to the distribution characteristics of each obstacle area; The processing module is also used to obtain the actual surgical path of the surgical robot in the current spinal surgery, and to perform feedback constraints on the obstacle influence degree according to the execution deviation between the actual surgical path and the initial surgical path, so as to obtain the confidence constraints of the obstacle avoidance strategy in the spinal surgery; The execution module is used to dynamically feedback the obstacle avoidance strategy of the surgical robot in the subsequent surgical path based on the confidence constraint of the obstacle avoidance strategy, so as to obtain the obstacle avoidance path of the surgical robot in spinal surgery.

[0013] In a third aspect, the present application provides a computer device, comprising a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the computer device executes the above-mentioned percutaneous spinal endoscopic surgery robot path planning method.

[0014] In a fourth aspect, the present application provides a computer-readable storage medium, in which instructions or codes are stored. When the instructions or codes are run on a computer, the computer implements the above-mentioned percutaneous spinal endoscopic surgery robot path planning method when executed.

[0015] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects: The present application provides a percutaneous spinal endoscopic surgical robot path planning method and system, which generates an initial surgical path for percutaneous spinal endoscopy through anatomical images of the spine, and monitors the tissue structure information of the percutaneous spine during the spinal surgery when the surgical robot performs spinal surgery according to the initial surgical path; determines the pixel offset characteristics of the percutaneous spine after tissue structure deformation in the current spinal surgery based on deformation data in the tissue structure information, determines various obstacle areas caused by tissue deformation in the percutaneous spine through the pixel offset characteristics and the current motion state of the surgical robot, and then predicts the obstacle influence of tissue deformation in the spinal surgery at the current surgical moment based on the distribution characteristics of each obstacle area; obtains the actual surgical path of the surgical robot in the current spinal surgery, and performs feedback constraints on the obstacle influence according to the execution deviation between the actual surgical path and the initial surgical path to obtain the confidence constraints of the obstacle avoidance strategy in the spinal surgery; and dynamically feeds back the obstacle avoidance strategy of the surgical robot in subsequent surgical paths based on the confidence constraints of the obstacle avoidance strategy to obtain the obstacle avoidance path of the surgical robot in the spinal surgery.

[0016] It can be seen that in this application, based on the confidence constraint of the obstacle avoidance strategy, the obstacle avoidance strategy of the surgical robot in the subsequent surgical path is dynamically fed back to obtain the obstacle avoidance path of the surgical robot in spinal surgery; first, the obstacle influence degree is determined to obtain the interference of obstacles caused by tissue deformation on the surgical path, which helps the surgical robot to perceive the dynamic changes in the surgical environment in real time, including potential obstacles caused by tissue deformation. Real-time prediction can provide key input for dynamic feedback regulation, so that the surgical robot can make timely adjustments to the path planning according to the changes in the current surgical scene, thereby avoiding potential path conflicts or surgical failures. The safety and accuracy of the surgical procedure are greatly improved; then, the influence of tissue deformation on the obstacle avoidance path in spinal surgery can be obtained by determining the confidence constraints. The obstacle influence and path planning adjustment requirements can be accurately quantified through a closed-loop feedback mechanism, so that the surgical robot can optimize the path planning strategy according to the actual surgical path, ensuring the accuracy and adaptability of the obstacle avoidance path, while reducing the intraoperative risks caused by improper path planning, thereby further improving the safety of percutaneous endoscopic spinal surgery; in summary, based on the above scheme, dynamic feedback adjustment of the obstacle avoidance path in the surgical robot can be realized, thereby improving the safety of percutaneous endoscopic spinal surgery. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0018] Figure 1 is an exemplary flow chart of a path planning method for percutaneous spinal endoscopic surgery robot according to some embodiments of the present application; Figure 2 is an optical principle diagram of a spinal endoscope according to some embodiments of the present application; Figure 3 is a schematic diagram of a process for determining confidence constraints according to some embodiments of the present application; Figure 4 is a schematic diagram of the structure of a percutaneous spinal endoscopic surgery robot path planning system according to some embodiments of the present application; Figure 5 It is a structural schematic diagram of a computer device for implementing a percutaneous spinal endoscopic surgery robot path planning method according to some embodiments of the present application. DETAILED DESCRIPTION

[0019] In order to better understand the technical solution of the present application, the technical solution of the present application will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0020] refer to Figure 1 , which is an exemplary flow chart of a percutaneous spinal endoscopic surgery robot path planning method according to some embodiments of the present application, and the percutaneous spinal endoscopic surgery robot path planning method mainly includes the following steps: In step 101, an initial percutaneous spinal endoscopy surgical path is generated through anatomical images of the spine, and when a surgical robot performs spinal surgery according to the initial surgical path, the tissue structure information of the percutaneous spine during the spinal surgery is monitored.

[0021] It should be noted that in the present application, the anatomical image represents a three-dimensional image of the anatomical structure of the spine and surrounding tissues; the initial surgical path represents the predetermined path to be executed by the surgical robot; and the tissue structure information represents the physical properties, positional relationships and deformation data of the spine and its surrounding tissues.

[0022] In specific implementation, computed tomography can be used to obtain a high-resolution three-dimensional image of the spine as an anatomical image of the spine, and then the contours and positions of bones, nerves, and blood vessels can be extracted through an image segmentation algorithm (for example, U-Net based on deep learning), so as to perform three-dimensional reconstruction of the segmented anatomical image and generate a complete anatomical model. A path planning algorithm (for example, Dijkstra algorithm) is used to generate a path from the needle entry point to the target position in the three-dimensional anatomical model, i.e., an initial surgical path. When the surgical robot performs spinal surgery according to the initial surgical path, ultrasonic imaging equipment can be used to monitor the tissue structure information of the percutaneous spine during spinal surgery.

[0023] In some embodiments, reference Figure 2 The figure is an optical principle diagram of a spinal endoscope according to some embodiments of the present application, and the figure shows the optical system of the microendoscope in the spinal endoscope, which allows doctors to obtain a clear internal view when performing spinal surgery.

[0024] First, the light enters from the large field of view on the left, which is the wide area that the endoscope can observe; then, the light passes through the objective lens, which collects and focuses the light to form the first imaging plane; the design of the objective lens is crucial because it directly affects the quality of the image and the breadth of the field of view; next, the light passes through three sets of image relay systems, each of which contains prisms or mirrors to change the direction of the light and ensure the correct alignment of the image; the image relay system allows the endoscope to provide high-quality images while maintaining its length and flexibility; during the image relay process, the light passes through the secondary, tertiary and quaternary imaging planes, and each imaging is a further focus and adjustment of the light to ensure the clarity and accuracy of the image; finally, the light passing through the eyepiece is guided by the adapter to the exit pupil, which is where the observer's eye receives the light; the design of the exit pupil must ensure the doctor's comfort and image clarity during observation.

[0025] The entire optical system is designed to provide high-definition images, allowing doctors to accurately observe the details inside the spine and perform precise surgical operations.

[0026] In step 102, the pixel offset characteristics of the percutaneous spine after the tissue structure is deformed in the current spinal surgery are determined based on the deformation data in the tissue structure information, and the various obstacle areas caused by the tissue deformation in the percutaneous spine are determined through the pixel offset characteristics and the current motion state of the surgical robot. Then, the obstacle impact of the tissue deformation in the spinal surgery at the current moment of the operation is predicted based on the distribution characteristics of each obstacle area.

[0027] In some embodiments, determining the pixel shift characteristics of the percutaneous spine after the tissue structure is deformed in the current spinal surgery according to the deformation data in the tissue structure information can be achieved by the following steps: Comparing the anatomical image of the spine with the tissue structure information to obtain a deformation field between the real-time image and the original image during spinal surgery; Dividing the deformation field into a plurality of local interest regions based on the regional range of the initial surgical path, and then determining deformation sub-features of tissue structures in each local interest region; The pixel offset features of the percutaneous spine after the tissue structure is deformed in the current spinal surgery are determined according to all the deformation sub-features.

[0028] It should be noted that in the present application, the pixel offset feature represents the degree of spatial offset caused by deformation of tissue structure during spinal surgery; the deformation field is a field quantity that describes the overall spatial distribution of deformation of tissue structure during surgery, and the deformation field includes the direction, amplitude and regional range of deformation; the deformation sub-feature reflects the deformation law of tissue structure in the local area of ​​interest.

[0029] In the specific implementation, first, the anatomical image of the spine and the tissue structure information can be aligned using an anatomical image registration algorithm (for example, the Demons algorithm based on mutual information), so that the different areas between the anatomical image of the spine and the tissue structure information can be used as the deformation field between the real-time image and the original image in spinal surgery; then, the impact distance of spinal surgery on the tissue structure can be preset based on the structural characteristics around the skinned spine and historical experience, and the range from the initial surgical path to the impact distance can be used as the regional range of the initial surgical path, and a space division model is initialized. The regional range of the initial surgical path is used as the range of interest in the space division model, and the deformation field is used as the division target in the space division model. The space division is used. The model divides the region of interest to ensure that the deformation characteristics of the tissue structure in each region of interest are similar. The region after the region of interest is divided can be used as a local region of interest, and multiple local regions of interest can be obtained. For each local region of interest, the deformation law of displacement amplitude, deformation gradient and stress-strain distribution in the local region of interest is extracted based on the feature descriptor (for example: tensor). The set of all deformation laws can be used as the deformation sub-feature of the local region of interest. The law included in the deformation sub-feature can be obtained in the above manner. The deformation sub-feature of the tissue structure in each local region of interest can be obtained; finally, the set of all deformation sub-features can be used as the pixel offset feature after the deformation of the tissue structure of the percutaneous spine in the current spinal surgery.

[0030] In some embodiments, the following steps may be used to determine various obstacle areas caused by tissue deformation in the percutaneous spinal column by using the pixel offset feature and the current motion state of the surgical robot: Obtain the current motion state of the surgical robot, and then predict the actual motion path of the surgical robot in the current motion state; Predicting the offset impact area in the actual motion path by using the pixel offset feature; dividing the deviation-affected region into a plurality of obstacle regions based on tissue characteristics in the percutaneous spine; Individual obstruction areas resulting from tissue deformation in the percutaneous spine are determined from all obstruction areas.

[0031] It should be noted that, in the present application, the obstacle area refers to the short-term path blocking tissue caused by tissue deformation during surgery; the current motion state is a collection of motion information describing the surgical robot at the current moment, and the current motion state includes the current position, direction, speed and acceleration; the motion path represents the expected motion trajectory of the surgical robot from the current position to the end position of the surgery; the offset impact area refers to the spatial range where the surgical path is interfered with by the offset caused by tissue deformation; the obstacle area represents the spatial range where there are similar characteristic obstructions in the surgical path.

[0032] In the specific implementation, first, the current motion state of the surgical robot can be obtained from the central console of the surgical robot, and the path interpolation algorithm (for example: path estimation based on Kalman filtering) can be used to generate the actual motion path of the surgical robot in the current motion state; secondly, a deformation transfer model based on finite element simulation is initialized, and each deformation sub-feature in the pixel offset feature is used as the deformation parameter of the tissue structure in the deformation transfer model. The deformation transfer model is used to predict the area affected by the deformation of the tissue structure in the actual motion path, and the predicted result can be used as the offset affected area in the actual motion path; then, the partition rules are defined by the tissue characteristics in the percutaneous spine, for example For example, areas with higher hardness are marked as non-crossable, and soft tissue areas are marked as low priority obstacle avoidance. The partitioning rules combined with the region growing method can be used to divide the offset affected area, so that the area where the difference between tissue characteristics in the percutaneous spine is within the preset difference threshold range is taken as an obstacle area, and multiple obstacle areas can be obtained; finally, for each obstacle area, an image processing algorithm (for example: region growing method) is used to identify the obstacle characteristics in the obstacle area, and the obstacle area carrying the obstacle characteristics can be taken as the obstacle area corresponding to the obstacle area. The obstacle area corresponding to each obstacle area can be obtained in the above manner, and the various obstacle areas caused by tissue deformation in the percutaneous spine can be obtained.

[0033] It should be noted that in the present application, the deformation transfer model is a model used to predict and describe the impact of tissue deformation on the surgical path. The technical principle of the deformation transfer model is based on the finite element analysis method, and the mechanical properties of the tissue (for example: elastic modulus) are used to model the spinal structure to simulate the deformation that may occur during surgery. The pixel offset feature is used as input. By comparing the original and deformed states of the tissue, the deformation transfer model can accurately predict the path offset area caused by tissue deformation. The output result of the model is the offset impact area in the actual motion path. Combined with the hardness and other characteristics of the spinal tissue, it can realize the dynamic division of the obstacle area, thereby providing accurate obstacle avoidance information for the path planning of the surgical robot.

[0034] In some embodiments, predicting the obstacle influence of tissue deformation in spinal surgery at the current surgery time based on the distribution characteristics of each obstacle area can be achieved by the following steps: Extract features from each obstacle area to obtain distribution features of each obstacle area; Determine the interference information of tissue deformation on spinal surgery through all distribution characteristics; Based on the obstacle interference information, a cross prediction is performed on the path interference in the initial surgical path to obtain the obstacle influence degree of tissue deformation in spinal surgery at the current surgical moment.

[0035] It should be noted that, in the present application, obstacle impact refers to the interference of obstacles caused by tissue deformation on the surgical path; distribution characteristics refer to the spatial distribution pattern of obstacles in the surgical area; obstacle interference information is an indicator to measure the degree of interference caused by obstacles on the surgical path.

[0036] In specific implementation, first, for each obstacle area, a feature extraction algorithm (for example, principal component analysis) can be used to extract the discrete degree of tissue distribution in the obstacle area, and the quantified value of the extracted tissue distribution law is used as the distribution feature of the obstacle area. The distribution features of each obstacle area can be obtained in the above manner; then, a mechanical analysis model can be used to superimpose and evaluate all distribution features with the initial surgical path, so as to evaluate the resistance interference of each obstacle area on the movement of the surgical robot, and the collection of the resistance interference can be used as the obstacle interference information of tissue deformation on spinal surgery; finally, a perturbation model is established using the initial surgical path, and each resistance interference in the obstacle interference information is used as the path interference degree of the corresponding area of ​​each obstacle area in the perturbation model, and the perturbation model is used to cross-compare the resistance interference of the path in different positions, and the quantitative result of the cross-comparison can be used as the obstacle influence degree of tissue deformation in spinal surgery at the current surgical moment.

[0037] In step 103, the actual surgical path of the surgical robot in the current spinal surgery is obtained, and the obstacle influence is feedback constrained according to the execution deviation between the actual surgical path and the initial surgical path to obtain the confidence constraint of the obstacle avoidance strategy in the spinal surgery.

[0038] In some embodiments, the actual surgical path of the surgical robot in the current spinal surgery is obtained. In specific implementation, a position sensor (for example, a laser sensor) installed on the surgical robot can be used to collect the real-time position of the movement of the robotic arm of the surgical robot every 1 second. The real-time position from the start of the spinal surgery to the current moment is connected in an orderly manner to obtain the actual surgical path of the surgical robot in the current spinal surgery. It should be noted that, in the present application, the actual surgical path is the path actually traversed by the surgical robot during the execution process.

[0039] In some embodiments, feedback constraints are applied to the obstacle influence degree according to the execution deviation between the actual surgical path and the initial surgical path to obtain the confidence constraints of the obstacle avoidance strategy in spinal surgery. Figure 3 As described above, the figure is a schematic diagram of a process for determining a confidence constraint in some embodiments of the present application. In this embodiment, the confidence constraint can be determined by the following steps: In step 1031, the execution deviation between the actual surgical path and the initial surgical path is determined by a path comparison algorithm; In step 1032, a feedback adjustment amount of the execution deviation on the actual surgical path is determined; In step 1033, the obstacle avoidance constraint amount of the obstacle influence degree on the actual surgical path is determined; In step 1034, the confidence constraint of the obstacle avoidance strategy in spinal surgery is determined based on the feedback adjustment amount and the deviation constraint amount.

[0040] It should be noted that in the present application, the confidence constraint represents the degree of constraint of tissue deformation on the obstacle avoidance path during spinal surgery; the execution deviation is an indicator that measures the degree of deviation between the surgical robot and the planned path in actual operation; the feedback adjustment amount represents the path adjustment amount caused by the deformation obstacle caused by the execution deviation; and the obstacle avoidance constraint amount is a quantitative value of the degree of restriction of the obstacle area on the path planning.

[0041] In specific implementation, first, a path comparison algorithm (for example, Hausdorff distance) can be used to match the actual surgical path and the initial surgical path point by point to obtain a quantified deviation value for each path point, and the set of all quantified deviation values ​​can be used as the execution deviation between the actual surgical path and the initial surgical path; secondly, the product of the mean of all quantified deviation values ​​in the execution deviation and the obstacle influence degree can be used as the feedback adjustment amount of the execution deviation on the obstacle influence degree; then, a path constraint model based on finite element simulation is initialized, the actual surgical path is used as the force boundary condition in the path constraint model, and the obstacle influence degree is used as the external disturbance factor in the path constraint model, so as to use the path constraint model to generate the obstacle avoidance constraint amount of the obstacle influence degree on the actual surgical path; finally, the product of the feedback adjustment amount and the deviation constraint amount can be used as the confidence constraint of the obstacle avoidance strategy in spinal surgery.

[0042] It should be noted that the path constraint model is a model used to constrain and adjust the motion path of the surgical robot. The path constraint model uses the mechanical properties of the tissue and the force boundary conditions to model the spinal surgery path. In the path constraint model, the actual surgical path is defined as the force boundary conditions, that is, the motion trajectory followed by the surgical robot during the execution process. At the same time, the obstacle influence is regarded as an external disturbance factor to simulate the impact of tissue structure deformation or environmental changes on the path. The path constraint model generates an obstacle avoidance constraint through a comprehensive analysis of the force boundary conditions and external disturbances, guiding the surgical robot to avoid obstacles and optimize path execution during the operation to ensure surgical accuracy and safety.

[0043] In step 104, the obstacle avoidance strategy of the surgical robot in the subsequent surgical path is dynamically fed back based on the confidence constraint of the obstacle avoidance strategy to obtain the obstacle avoidance path of the surgical robot in spinal surgery.

[0044] In some embodiments, the obstacle avoidance strategy of the surgical robot in the subsequent surgical path is dynamically fed back based on the confidence constraint of the obstacle avoidance strategy, and the obstacle avoidance path of the surgical robot in spinal surgery can be obtained by the following steps: Initialize a dynamic obstacle avoidance model for a path planning algorithm; Determining the dynamic constraint conditions of the dynamic obstacle avoidance model through the confidence constraints of the obstacle avoidance strategy; Determine the target optimization function of the dynamic obstacle avoidance model through the obstacle avoidance strategy of the initial surgical path; The dynamic obstacle avoidance model is used to feedback-adjust the obstacle avoidance strategy of the surgical robot in the subsequent surgical path, and the obstacle avoidance path of the surgical robot in spinal surgery is obtained.

[0045] It should be noted that in the present application, the obstacle avoidance path refers to the path of the surgical robot to avoid obstacles during execution; the dynamic obstacle avoidance model is a model used to adjust the path of the surgical robot in real time to cope with external disturbances and changes; the dynamic constraint conditions represent the restriction conditions for obstacle avoidance in the path planning of the surgical robot; and the objective optimization function is a mathematical formula that describes the obstacle avoidance characteristics of the surgical robot in path planning.

[0046] In the specific implementation, first, a path planning algorithm suitable for the surgical scene (for example, an RRT algorithm based on optimization) is selected as the core framework of the model to initialize a dynamic obstacle avoidance model; secondly, the confidence constraint is used as the dynamic constraint condition of the dynamic obstacle avoidance model to reflect the impact of environmental changes on the path; then, the initial optimization function in the dynamic obstacle avoidance model is obtained, and the obstacle avoidance strategy of the initial surgical path can be obtained from the path planning algorithm that generates the initial surgical path, so that the minimized path length, smoothness threshold and obstacle avoidance condition in the obstacle avoidance strategy are used as the corresponding parameter values ​​in the initial optimization function, and the target optimization function of the dynamic obstacle avoidance model can be obtained, which can ensure the optimality of the path. The optimization function is continuously updated in the dynamic obstacle avoidance model to adapt to the real-time changes of obstacles during the operation; finally, the dynamic obstacle avoidance model is used to feedback-adjust the obstacle avoidance strategy of the surgical robot in the subsequent surgical path, and the path generated by the obstacle avoidance strategy after feedback adjustment can be used as the obstacle avoidance path of the surgical robot in spinal surgery. The obstacle avoidance strategy adjusted by feedback adjustment updates the path of the surgical robot in real time, ensuring that the robot can accurately and safely avoid obstacles and optimize the motion trajectory, thereby realizing accurate navigation and operation of spinal surgery.

[0047] In addition, in another aspect of the present application, in some embodiments, the present application provides a percutaneous spinal endoscopic surgery robot path planning system, referring to Figure 4, which is a schematic diagram of the structure of a percutaneous spinal endoscopic surgery robot path planning system according to some embodiments of the present application. The percutaneous spinal endoscopic surgery robot path planning system includes: a monitoring module 201, a processing module 202 and an execution module 203, which are described as follows: Monitoring module 201, in the present application, monitoring module 201 is mainly used to generate an initial surgical path of percutaneous spinal endoscopy through anatomical images of the spine, and to monitor the tissue structure information of the percutaneous spine during the spinal surgery when the surgical robot performs the spinal surgery according to the initial surgical path; The processing module 202 in the present application is used to determine the pixel offset characteristics of the percutaneous spine after the tissue structure is deformed in the current spinal surgery according to the deformation data in the tissue structure information, determine the various obstacle areas caused by the tissue deformation in the percutaneous spine through the pixel offset characteristics and the current motion state of the surgical robot, and then predict the obstacle influence degree of the tissue deformation in the spinal surgery at the current surgery time according to the distribution characteristics of each obstacle area; It should be noted that the processing module 202 is also used to obtain the actual surgical path of the surgical robot in the current spinal surgery, and to perform feedback constraints on the obstacle influence according to the execution deviation between the actual surgical path and the initial surgical path, so as to obtain the confidence constraints of the obstacle avoidance strategy in the spinal surgery; Execution module 203. In the present application, execution module 203 is mainly used to dynamically feedback the obstacle avoidance strategy of the surgical robot in the subsequent surgical path based on the confidence constraint of the obstacle avoidance strategy, so as to obtain the obstacle avoidance path of the surgical robot in spinal surgery.

[0048] The above describes in detail the examples of the percutaneous spinal endoscopic surgery robot path planning method and system provided by the embodiments of the present application. It can be understood that in order to realize the above functions, the corresponding device includes hardware structures and / or software modules corresponding to the execution of each function. It should be easily appreciated by those skilled in the art that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present application.

[0049] In some embodiments, the present application also provides a computer device, which includes a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the computer device executes the above-mentioned percutaneous spinal endoscopic surgery robot path planning method.

[0050] In some embodiments, reference Figure 5 , the dotted line in the figure indicates that the unit or module is optional, and the figure is a schematic diagram of the structure of a computer device for implementing a path planning method for a percutaneous spinal endoscopic surgery robot according to an embodiment of the present application. The path planning method for a percutaneous spinal endoscopic surgery robot described in the above embodiment can be Figure 5 The computer device shown in the figure is implemented, and the computer device includes at least one processor 301, a memory 302 and at least one communication unit 305. The computer device can be a terminal device, a server or a chip.

[0051] The processor 301 may be a general-purpose processor or a special-purpose processor. For example, the processor 301 may be a central processing unit (CPU), which may be used to control the computer device, execute software programs, and process data of the software programs. The computer device may also include a communication unit 305 to implement signal input (reception) and output (transmission).

[0052] For example, the computer device may be a chip, the communication unit 305 may be an input and / or output circuit of the chip, or the communication unit 305 may be a communication interface of the chip, and the chip may be a component of a terminal device, a network device, or other devices.

[0053] For another example, the computer device may be a terminal device or a server, and the communication unit 305 may be a transceiver of the terminal device or the server, or the communication unit 305 may be a transceiver circuit of the terminal device or the server.

[0054] The computer device may include one or more memories 302, on which a program 304 is stored. The program 304 can be executed by the processor 301 to generate instructions 303, so that the processor 301 performs the method described in the above method embodiment according to the instructions 303. Optionally, data (such as a target audit model) can also be stored in the memory 302. Optionally, the processor 301 can also read the data stored in the memory 302, and the data can be stored at the same storage address as the program 304, or the data can be stored at a different storage address from the program 304.

[0055] The processor 301 and the memory 302 may be provided separately or integrated together, for example, integrated on a system on chip (SOC) of the terminal device.

[0056] It should be understood that each step of the above method embodiment can be completed by a hardware-based logic circuit or software-based instructions in the processor 301. The processor 301 can be a CPU, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, such as discrete gates, transistor logic devices, or discrete hardware components.

[0057] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0058] For example, in some embodiments, the present application also provides a computer-readable storage medium, in which instructions or codes are stored. When the instructions or codes are run on a computer, the computer implements the above-mentioned percutaneous spinal endoscopic surgery robot path planning method when executed.

[0059] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0060] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A robot path planning method for percutaneous spinal endoscopic surgery, characterized in that: The steps include: Generate the initial surgical path of percutaneous spinal endoscopy through anatomical images of the spine, and monitor the tissue structure information of the percutaneous spine during spinal surgery when the surgical robot performs spinal surgery according to the initial surgical path; Determine pixel offset features of the percutaneous spine after the tissue structure is deformed in the current spinal surgery according to the deformation data in the tissue structure information, determine various obstacle areas caused by the tissue deformation in the percutaneous spine through the pixel offset features and the current motion state of the surgical robot, and then predict the obstacle influence degree of the tissue deformation in the spinal surgery at the current surgery moment according to the distribution characteristics of each obstacle area; Acquire the actual surgical path of the surgical robot in the current spinal surgery, perform feedback constraints on the obstacle influence according to the execution deviation between the actual surgical path and the initial surgical path, and obtain the confidence constraints of the obstacle avoidance strategy in the spinal surgery; Based on the confidence constraint of the obstacle avoidance strategy, the obstacle avoidance strategy of the surgical robot in the subsequent surgical path is dynamically fed back to obtain the obstacle avoidance path of the surgical robot in spinal surgery.

2. The method according to claim 1, characterized in that Determining the pixel shift characteristics of the percutaneous spine after the tissue structure is deformed in the current spinal surgery according to the deformation data in the tissue structure information specifically includes: Comparing the anatomical image of the spine with the tissue structure information to obtain a deformation field between the real-time image and the original image during spinal surgery; Dividing the deformation field into a plurality of local interest regions based on the regional range of the initial surgical path, and then determining deformation sub-features of tissue structures in each local interest region; The pixel offset features of the percutaneous spine after the tissue structure is deformed in the current spinal surgery are determined according to all the deformation sub-features.

3. The method according to claim 1, characterized in that The various obstacle areas caused by tissue deformation in the percutaneous spine are determined by the pixel offset feature and the current motion state of the surgical robot, specifically including: Obtain the current motion state of the surgical robot, and then predict the actual motion path of the surgical robot in the current motion state; Predicting the offset impact area in the actual motion path by using the pixel offset feature; dividing the deviation-affected region into a plurality of obstacle regions based on tissue characteristics in the percutaneous spine; Individual obstruction areas resulting from tissue deformation in the percutaneous spine are determined from all obstruction areas.

4. The method according to claim 1, characterized in that The influence of tissue deformation on the current operation time during spinal surgery is predicted by the distribution characteristics of each obstacle area, including: Extract features from each obstacle area to obtain distribution features of each obstacle area; Determine the interference information of tissue deformation on spinal surgery through all distribution characteristics; Based on the obstacle interference information, a cross prediction is performed on the path interference in the initial surgical path to obtain the obstacle influence degree of tissue deformation in spinal surgery at the current surgical moment.

5. The method according to claim 1, characterized in that Feedback constraints are applied to the obstacle influence degree according to the execution deviation between the actual surgical path and the initial surgical path, and the confidence constraints of the obstacle avoidance strategy in spinal surgery are obtained, which specifically include: Determining the execution deviation between the actual surgical path and the initial surgical path by a path comparison algorithm; Determining a feedback adjustment amount of the execution deviation on an actual surgical path; Determining the obstacle avoidance constraint amount of the obstacle influence on the actual surgical path; The confidence constraint of the obstacle avoidance strategy in spinal surgery is determined based on the feedback adjustment amount and the deviation constraint amount.

6. The method according to claim 1, characterized in that Based on the confidence constraint of the obstacle avoidance strategy, the obstacle avoidance strategy of the surgical robot in the subsequent surgical path is dynamically fed back to obtain the obstacle avoidance path of the surgical robot in spinal surgery, which specifically includes: Initialize a dynamic obstacle avoidance model for a path planning algorithm; Determining the dynamic constraint conditions of the dynamic obstacle avoidance model through the confidence constraints of the obstacle avoidance strategy; Determine the target optimization function of the dynamic obstacle avoidance model through the obstacle avoidance strategy of the initial surgical path; The dynamic obstacle avoidance model is used to feedback-adjust the obstacle avoidance strategy of the surgical robot in the subsequent surgical path, and the obstacle avoidance path of the surgical robot in spinal surgery is obtained.

7. The method according to claim 1, characterized in that Ultrasound imaging is used to monitor the structural information of the spine during percutaneous spinal surgery.

8. A percutaneous spinal endoscopic surgery robot path planning system, characterized in that: include: A monitoring module, used to generate an initial surgical path for percutaneous spinal endoscopy through anatomical images of the spine, and to monitor the tissue structure information of the percutaneous spine during the spinal surgery when the surgical robot performs the spinal surgery according to the initial surgical path; A processing module, for determining pixel offset features of the percutaneous spine after the tissue structure is deformed in the current spinal surgery according to the deformation data in the tissue structure information, determining various obstacle areas caused by the tissue deformation in the percutaneous spine through the pixel offset features and the current motion state of the surgical robot, and then predicting the obstacle influence degree of the tissue deformation in the spinal surgery at the current surgery moment according to the distribution characteristics of each obstacle area; The processing module is also used to obtain the actual surgical path of the surgical robot in the current spinal surgery, and to perform feedback constraints on the obstacle influence degree according to the execution deviation between the actual surgical path and the initial surgical path, so as to obtain the confidence constraints of the obstacle avoidance strategy in the spinal surgery; The execution module is used to dynamically feedback the obstacle avoidance strategy of the surgical robot in the subsequent surgical path based on the confidence constraint of the obstacle avoidance strategy, so as to obtain the obstacle avoidance path of the surgical robot in spinal surgery.

9. A computer device, characterized in that: The computer device includes a memory and a processor, the memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the computer device executes the percutaneous spinal endoscopic surgery robot path planning method described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions or codes, and when the instructions or codes are executed on a computer, the computer implements the percutaneous spinal endoscopic surgery robot path planning method as described in any one of claims 1 to 7.

Images