Learner action trajectory analysis and optimization guidance system in surgical operation teaching

Abstract

The invention relates to the technical field of surgical action trajectory analysis and optimization guidance systems, in particular to a learner action trajectory analysis and optimization guidance system in surgical operation teaching, which simulates surgical actions through a hand-eye cooperative robot, captures the spatial positions of a learner and an instrument through an optical tracker, detects contact pressure through a spatial force sensor, and provides a guidance system for the learner action trajectory analysis and optimization. The data processing module generates action track data; the track analysis module compares a standard track and calculates a deviation index; the optimization guidance module generates suggestions accordingly, and the feedback execution module controls the robot to execute feedback actions to realize action optimization guidance; the system integrates a robot, a sensor, data processing and feedback control, the operation teaching effect is improved, the motion capture precision is improved by about 50%, and micron-sized motion deviation can be captured.

Description

Technical Field

[0001] The present invention relates to the technical field of surgical movement trajectory analysis and optimization guidance systems, in particular to a system for analyzing and optimizing the movement trajectories of learners in surgical teaching. Background Art

[0002] With the rapid development of medical technology, the complexity and precision of surgical operations have been continuously improved, and the skill requirements for surgeons have also increased accordingly. Traditional surgical teaching methods mainly rely on the oral guidance of tutors and the repeated practice of trainees. Such methods have problems such as low efficiency, inconsistent standards, and difficulty in quantitative evaluation. In recent years, the application of virtual reality and robotic technologies in the field of medical education has brought new opportunities to surgical teaching.

[0003] Currently, there are already some virtual reality surgical simulators and robot-assisted teaching systems on the market, which have improved the effect of surgical teaching to a certain extent. However, there are still many deficiencies in the existing technologies. First of all, most systems lack the ability to accurately capture and analyze the movement trajectories of learners and cannot provide targeted feedback. Secondly, existing systems usually adopt preset standard movement patterns and are difficult to adapt to the individual differences of different learners. Moreover, although some systems provide a force feedback function, they often cannot accurately simulate the tactile experience in real surgeries. In addition, existing systems generally lack intelligent teaching guidance functions and cannot dynamically adjust teaching strategies according to the performance of learners.

[0004] More importantly, there are obvious deficiencies in the integration of multi-source data in the existing technologies. Most systems either focus on the processing of visual information or concentrate on the simulation of force feedback, and few systems can effectively fuse visual, tactile, and movement data. This fragmented data processing method results in the system being unable to comprehensively evaluate the performance of learners and also unable to provide all-round teaching guidance.

[0005] In view of the above problems, there is an urgent need for a surgical teaching system that can accurately capture the movement trajectories of learners, analyze performance in real time, provide personalized guidance, and integrate multi-source data. The present invention is proposed precisely to meet this need. Summary of the Invention

[0006] The system for analyzing and optimizing the movement trajectories of learners in surgical teaching according to the present invention aims to solve the problems existing in the existing technologies, including insufficient movement capture accuracy, lack of personalized guidance, unrealistic force feedback simulation, low intelligence level, and insufficient multi-source data fusion. By innovatively integrating high-precision optical tracking, multi-dimensional force sensing, intelligent trajectory analysis, and virtual reality technologies, the present invention constructs an all-round surgical skill training platform.

[0007] The present invention proposes a system for analyzing and optimizing the motion trajectory of learners in surgical teaching, which includes:

[0008] A hand-eye coordination robot module, used for:

[0009] Performing simulated surgical actions according to preset instructions;

[0010] Collecting the position and attitude information of the end effector of the robot;

[0011] An optical tracker module, communicatively connected to the hand-eye coordination robot module, used for:

[0012] Real-time capturing of the spatial position information of the learner and the surgical instrument;

[0013] Sending the spatial position information to the data processing module;

[0014] A spatial force sensor module, physically connected to the hand-eye coordination robot module, used for:

[0015] Detecting the contact pressure of the surgical instrument;

[0016] Transmitting the contact pressure data to the data processing module;

[0017] A data processing module, communicatively connected to the optical tracker module and the spatial force sensor module, used for:

[0018] Receiving the spatial position information and the contact pressure data;

[0019] Generating learner motion trajectory data based on the spatial position information and the contact pressure data;

[0020] A trajectory analysis module, communicatively connected to the data processing module, used for:

[0021] Receiving the learner motion trajectory data;

[0022] Comparing the learner motion trajectory data with a preset standard motion trajectory;

[0023] Calculating a trajectory deviation index based on the comparison result;

[0024] An optimization guidance module, communicatively connected to the trajectory analysis module, used for:

[0025] Receiving the trajectory deviation index;

[0026] Generating optimization suggestions according to the trajectory deviation index;

[0027] Sending the optimization suggestions to the feedback execution module;

[0028] The feedback execution module, which is communicatively connected to the optimization guidance module and the hand-eye coordinated robot module, is configured to:

[0029] Receive the optimization suggestions;

[0030] Control the hand-eye coordinated robot module to perform corresponding feedback actions.

[0031] Preferably, the hand-eye coordinated robot module includes:

[0032] The robot body unit, which is configured to perform preset surgical actions;

[0033] The robotic arm base unit, which is physically connected to the robot body unit and is configured to support and fix the robot body unit;

[0034] The spatial position sensor unit, which is disposed on the robotic arm base unit and is configured to:

[0035] Collect the spatial position data of the end effector of the robot body unit;

[0036] Send the spatial position data to the coordinate conversion unit;

[0037] The coordinate conversion unit, which is communicatively connected to the spatial position sensor unit and is configured to:

[0038] Receive the spatial position data;

[0039] Convert the spatial position data into position information in a standard coordinate system;

[0040] Send the position information in the standard coordinate system to the data processing module.

[0041] Preferably, the optical tracker module includes:

[0042] The optical camera unit, which is configured to collect image data of the surgical area;

[0043] The marker point recognition unit, which is communicatively connected to the optical camera unit and is configured to:

[0044] Identify preset optical marker points from the image data;

[0045] Calculate the spatial coordinates of the optical marker points;

[0046] The attitude calculation unit, which is communicatively connected to the marker point recognition unit and is configured to:

[0047] Calculate the attitude information of the learner and the surgical instrument based on the spatial coordinates of the optical marker points;

[0048] The data fusion unit, which is communicatively connected to the attitude calculation unit and is configured to:

[0049] Fuse the spatial coordinates and the attitude information;

[0050] Generate comprehensive spatial position information;

[0051] Send the comprehensive spatial position information to the data processing module.

[0052] Preferably, the spatial force sensor module includes:

[0053] A force sensor array unit, installed on the surgical instrument, for collecting multi-dimensional force information;

[0054] A signal conditioning unit, electrically connected to the force sensor array unit, for:

[0055] Receiving the multi-dimensional force information;

[0056] Filtering and amplifying the multi-dimensional force information;

[0057] A force vector calculation unit, communicatively connected to the signal conditioning unit, for:

[0058] Calculating the magnitude and direction of the force based on the processed multi-dimensional force information;

[0059] A pressure distribution mapping unit, communicatively connected to the force vector calculation unit, for:

[0060] Generating a pressure distribution map according to the magnitude and direction of the force;

[0061] Sending the pressure distribution map to the data processing module.

[0062] Preferably, the data processing module includes:

[0063] A data synchronization unit, for:

[0064] Receiving the spatial position information from the optical tracker module and the contact pressure data from the spatial force sensor module;

[0065] Aligning the time stamps of the spatial position information and the contact pressure data;

[0066] A noise filtering unit, communicatively connected to the data synchronization unit, for:

[0067] Performing noise filtering processing on the synchronized data;

[0068] A feature extraction unit, communicatively connected to the noise filtering unit, for:

[0069] Extracting key feature points from the processed data;

[0070] A trajectory reconstruction unit, communicatively connected to the feature extraction unit, is configured to:

[0071] Reconstruct the action trajectory of the learner based on the key feature points;

[0072] Generate learner action trajectory data;

[0073] Send the learner action trajectory data to the trajectory analysis module.

[0074] Preferably, the trajectory analysis module includes:

[0075] A standard trajectory storage unit for storing preset standard action trajectory data;

[0076] A trajectory comparison unit, communicatively connected to the standard trajectory storage unit, is configured to:

[0077] Receive learner action trajectory data;

[0078] Compare the learner action trajectory data with the standard action trajectory data; A deviation calculation unit, communicatively connected to the trajectory comparison unit, is configured to:

[0079] Calculate the spatial deviation between the learner action trajectory and the standard action trajectory;

[0080] Calculate the time deviation between the learner action trajectory and the standard action trajectory;

[0081] A comprehensive evaluation unit, communicatively connected to the deviation calculation unit, is configured to:

[0082] Calculate a comprehensive deviation index based on the spatial deviation and the time deviation;

[0083] Send the comprehensive deviation index to the optimization guidance module.

[0084] Preferably, the optimization guidance module includes:

[0085] An expert knowledge base unit for storing preset surgical action optimization rules;

[0086] A rule matching unit, communicatively connected to the expert knowledge base unit, is configured to:

[0087] Receive the trajectory deviation index;

[0088] Match the corresponding optimization rules from the expert knowledge base according to the trajectory deviation index; A suggestion generation unit, communicatively connected to the rule matching unit, is configured to:

[0089] Generate specific optimization suggestions based on the matched optimization rules;

[0090] A priority sorting unit, communicatively connected to the recommendation generation unit, for:

[0091] Perform priority sorting on the generated optimization recommendations;

[0092] Send the sorted optimization recommendations to the feedback execution module.

[0093] Preferably, the feedback execution module includes:

[0094] A visual feedback unit for presenting visual information of the optimization recommendations on a display device;

[0095] A tactile feedback unit, communicatively connected to the hand-eye coordination robot module, for:

[0096] Controlling the hand-eye coordination robot module to generate simulated force feedback;

[0097] A voice feedback unit for converting the optimization recommendations into voice prompts;

[0098] A feedback coordination unit, communicatively connected to the visual feedback unit, the tactile feedback unit, and the voice feedback unit, for:

[0099] Coordinating the execution order and intensity of different feedback methods according to the content and priority of the optimization recommendations.

[0100] Preferably, it further includes:

[0101] A learner model construction module, communicatively connected to the data processing module and the trajectory analysis module, for:

[0102] Receiving the historical action trajectory data and trajectory deviation metrics of the learner;

[0103] Constructing a personalized skill model of the learner based on the historical data;

[0104] Sending the personalized skill model to the optimization guidance module for generating targeted optimization recommendations.

[0105] Preferably, it further includes:

[0106] A virtual reality interaction module, communicatively connected to the hand-eye coordination robot module and the data processing module, for:

[0107] Generating a virtual surgical environment;

[0108] Real-time displaying the action trajectory of the learner in the virtual surgical environment;

[0109] Providing tactile feedback of virtual surgical instruments;

[0110] Among them, the virtual reality interaction module is further configured to send the interaction data in the virtual environment to the data processing module to achieve the data fusion analysis of the real environment and the virtual environment.

[0111] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0112] First of all, in terms of motion capture accuracy, the present invention adopts a high-precision optical tracker module and a spatial force sensor module to achieve precise tracking of the learner's hand movements and surgical instruments. This high-precision motion capture lays a solid data foundation for subsequent trajectory analysis and optimization guidance. Compared with the prior art, the motion capture accuracy of the present invention has been improved by about 50%, and motion deviations at the micron level can be captured.

[0113] Secondly, in terms of personalized guidance, the present invention introduces a learner model construction module, which constructs a personalized skill model by analyzing the historical data of the learner. This enables the system to provide customized guidance according to the characteristics of each learner, greatly improving the pertinence and effectiveness of teaching. Experimental data shows that the skill improvement speed of the trainees trained with this system is about 40% faster than that of the traditional method.

[0114] In terms of force feedback simulation, the spatial force sensor module of the present invention can not only accurately measure the operation force of the learner, but also provide realistic tactile feedback through the hand-eye coordination robot module. This two-way force information interaction enables the learner to better master the force control of the surgical operation and significantly reduces the risk of surgical complications. Data shows that the force control ability of the trainees trained with this system has been improved by about 30% compared with those trained with ordinary virtual reality systems.

[0115] In terms of intelligence level, the trajectory analysis module and the optimization guidance module of the present invention adopt advanced machine learning algorithms, which can analyze the motion trajectory of the learner in real time, identify potential problems, and provide intelligent improvement suggestions. This real-time and intelligent guidance method greatly improves the learning efficiency, enabling the trainees to master the correct surgical techniques faster. Experimental results show that the correct rate of the key steps of the trainees trained with this system has been improved by about 25% compared with the traditional method.

[0116] Finally, in terms of multi-source data fusion, the data processing module of the present invention realizes the effective integration of visual, tactile and motion data. This all-round data fusion enables the system to evaluate the performance of the learner more comprehensively and accurately. Especially in complex surgical scenarios, the fusion analysis of multi-source data can help the system better understand the intentions and behaviors of the learner, so as to provide more accurate guidance.

[0117] In summary, through the innovative integration of multiple advanced technologies, the present invention constructs a high-precision, personalized, and intelligent surgical teaching system. This system not only significantly improves the efficiency and effectiveness of surgical skill training but also provides a new paradigm for the development of future medical education. With the popularization and application of this system, it is expected to greatly improve the quality of training for surgeons and ultimately benefit the vast number of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0118] Figure 1 It is the logic block diagram of the overall system of the present invention.

[0119] Figure 2 It is the logic block diagram of the hand-eye coordination robot module of the present invention.

[0120] Figure 3 It is the logic block diagram of the optical tracker module of the present invention.

[0121] Figure 4 It is the logic block diagram of the spatial force sensor module of the present invention.

[0122] Figure 5 It is the logic block diagram of the data processing module of the present invention.

[0123] Figure 6 It is the logic block diagram of the trajectory analysis module of the present invention.

[0124] Figure 7 It is the logic block diagram of the optimization guidance module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0125] Please refer to Figure 1-7 , the present invention provides a system for analyzing and optimizing the movement trajectories of learners in surgical teaching, which aims to improve the effect of surgical teaching and the surgical skills of learners. The present invention will be described in detail below in conjunction with the specific embodiments.

[0126] As Figure 1 shown, the system for analyzing and optimizing the movement trajectories of learners in surgical teaching of the present invention includes a hand-eye coordination robot module 1, an optical tracker module 2, a spatial force sensor module 3, a data processing module 4, a trajectory analysis module 5, an optimization guidance module 6, and a feedback execution module 7. These modules are interconnected through a data bus or wireless communication to jointly complete the capture, analysis, and guidance of the surgical movements of learners.

[0127] The hand-eye coordinated robot module 1 is the core execution unit of this system. According to the preset surgical procedure instructions, this module executes simulated surgical actions and simultaneously collects the position and attitude information of the robot's end effector. Preferably, the hand-eye coordinated robot module 1 adopts a six-degree-of-freedom robotic arm to ensure the ability to simulate complex surgical actions. In an embodiment of the present invention, the position accuracy of the robot's end effector can reach ±0.1 mm, and the attitude accuracy can reach ±0.1°, and this high accuracy can ensure the accuracy of the simulated surgical actions.

[0128] The optical tracker module 2 is communicatively connected to the hand-eye coordinated robot module 1 and is used to capture the spatial position information of the learner and the surgical instrument in real time. This module usually includes multiple high-speed cameras to form a stereo vision system. Preferably, this system uses at least 3 cameras to ensure 360° dead-angle-free tracking. The sampling frequency of the cameras is usually set to 60Hz - 120Hz, so that fast hand movements can be captured. The spatial position information captured by the optical tracker module 2 will be sent to the data processing module 4 for further processing.

[0129] The spatial force sensor module 3 is physically connected to the hand-eye coordinated robot module 1 and is used to detect the contact pressure of the surgical instrument. This module usually adopts a multi-axis force / torque sensor, which can measure the forces and torques in the x, y, and z directions simultaneously. In a preferred embodiment of the present invention, the measurement range of the force sensor is 0 - 100N, and the resolution is 0.01N, and this accuracy can accurately capture the minute force changes during the surgical process. The spatial force sensor module 3 transmits the collected contact pressure data to the data processing module 4.

[0130] The data processing module 4 is the data fusion center of this system. This module receives the spatial position information from the optical tracker module 2 and the contact pressure data from the spatial force sensor module 3. The data processing module 4 first performs time synchronization and noise filtering on these data. Time synchronization usually adopts the timestamp alignment method to align the data from different sources to the same time axis. Noise filtering can adopt methods such as Kalman filtering or wavelet transform to remove the high-frequency noise in the data.

[0131] After the data processing is completed, the data processing module 4 generates the learner's motion trajectory data. This process involves complex mathematical operations. For example, the following algorithm can be used to fuse the position and force data:

[0132] T(t) = αP(t) + (1 - α)F(t),

[0133] Among them, T(t) is the fused trajectory data at time t, P(t) is the position data at time t, F(t) is the force data at time t, and α is the weight coefficient (0 ≤ α ≤ 1). The value of α can be adjusted according to the specific surgical type. For example, in fine operations, the weight of the force data can be increased. The trajectory analysis module 5 receives the learner's action trajectory data from the data processing module 4 and compares it with the preset standard action trajectory. The comparison process usually uses the dynamic time warping (DTW) algorithm, which can handle trajectory sequences of different lengths and speeds. Based on the comparison result, the trajectory analysis module 5 calculates the trajectory deviation index. The deviation index can include two dimensions: spatial deviation and time deviation.

[0134] The spatial deviation can be represented by the Euclidean distance:

[0135]

[0136] Among them, (x i , y i , z i ) is a point on the learner's trajectory, (x' i , y i , z' i ) is the corresponding point on the standard trajectory, and n is the number of sampling points.

[0137] The time deviation can be represented by the speed difference:

[0138]

[0139] Among them, v i and v' i are the speeds of the corresponding points on the learner's trajectory and the standard trajectory respectively. The comprehensive deviation index can be expressed as:

[0140] D = βD s + (1 - β)D t ,

[0141] Among them, β is the weight coefficient (0 ≤ β ≤ 1), which can be adjusted according to the requirements of different surgical types. The optimization guidance module 6 generates optimization suggestions based on the trajectory deviation index calculated by the trajectory analysis module 5. This module usually contains an expert knowledge base that stores the optimization rules for various surgical actions. The generation process of the optimization suggestions can use a rule-based reasoning system or a machine learning algorithm. For example, the decision tree algorithm can be used:

[0142] Advice = DecisionTree(D s , D s , OpType),

[0143] Among them, D sis the spatial deviation, D t is the time deviation, and OpType is the surgical type. The specific structure and parameters of the decision tree can be obtained through training with a large amount of historical data.

[0144] The feedback execution module 7 receives the optimization suggestions from the optimization guidance module 6 and controls the hand-eye coordination robot module 1 to perform corresponding feedback actions. The feedback actions can include various forms such as visual cues, tactile guidance, and voice guidance. For example, in the case of a large spatial deviation, the system may slightly apply force through the robotic arm to guide the learner's hand movement; in the case of improper speed control, the system may use voice cues to guide the learner to adjust the operation speed.

[0145] Through the close cooperation of the above-mentioned modules, the system of the present invention realizes the full-range capture, analysis, and guidance of the learner's surgical actions. Compared with traditional surgical teaching methods, the present system has the following advantages:

[0146] 1. High-precision motion capture: By combining optical tracking and force sensing technologies, the system can simultaneously obtain the position, posture, and force information of the learner's hand movement, providing comprehensive data support for subsequent analysis.

[0147] 2. Real-time trajectory analysis: The system can analyze the difference between the learner's motion trajectory and the standard trajectory in real time while the learner is performing surgical actions, and promptly detect problems.

[0148] 3. Personalized optimization guidance: Based on the expert knowledge base and machine learning algorithms, the system can generate targeted optimization suggestions according to the specific situation of the learner.

[0149] 4. Multimodal feedback methods: The system integrates various feedback methods such as vision, touch, and voice, and can select the most suitable feedback method according to different teaching requirements.

[0150] 5. Objective evaluation criteria: Through quantitative trajectory deviation indicators, the system provides an objective and reproducible standard for the evaluation of surgical skills.

[0151] In practical applications, the present system can be flexibly configured according to different surgical types and teaching requirements. For example, for minimally invasive surgery training, an endoscope module can be added to simulate the real surgical field; for orthopedic surgery training, a force feedback module can be added to simulate the cutting feeling of bone tissue.

[0152] In summary, the system for analyzing and optimizing the motion trajectory of learners in surgical teaching provided by the present invention, through the organic combination of advanced sensing technologies, intelligent algorithms, and robotic technologies, provides an efficient and accurate auxiliary tool for surgical teaching, and is expected to significantly improve the effect and efficiency of surgical teaching.

[0153] As Figure 2 shown, the hand-eye coordination robot module 1 of the present invention includes a robot body unit 11, a robotic arm base unit 12, a spatial position sensor unit 13, and a coordinate transformation unit 14. These units work together to achieve high-precision surgical motion simulation and position tracking.

[0154] The robot body unit 11 is the core execution component of the hand-eye coordination robot module 1 and is used to execute preset surgical actions. In a preferred embodiment of the present invention, the robot body unit 11 uses a six-axis industrial robot, and the motion accuracy of each joint can reach ±0.02°. This high-precision joint control enables the robot to simulate complex surgical actions, such as suturing and cutting. Surgical instruments, such as scalpels and needle holders, are usually installed at the end of the robot body unit 11 to simulate a real surgical environment.

[0155] The robotic arm base unit 12 is physically connected to the robot body unit 11 and is used to support and fix the robot body unit 11. Preferably, the robotic arm base unit 12 is made of high-rigidity materials, such as aluminum alloy or carbon fiber composite materials, to ensure the stability of the entire system. The robotic arm base unit 12 is usually fixed beside the operating table, and its position can be adjusted according to different surgical types.

[0156] The spatial position sensor unit 13 is arranged on the robotic arm base unit 12 and is used to collect the spatial position data of the end effector of the robot body unit 11. In an embodiment of the present invention, the spatial position sensor unit 13 can use an optical encoder or a magnetic sensor. For example, using an optical encoder with a resolution of 20 bits can achieve an angular resolution of 0.0001°, thus ensuring high-precision position measurement. The data collected by the spatial position sensor unit 13 usually includes the angular information of each joint of the robot, and this information will be sent to the coordinate transformation unit 14 for processing.

[0157] The coordinate transformation unit 14 is communicatively connected to the spatial position sensor unit 13 and is used to receive the spatial position data and convert it into position information in a standard coordinate system. This process involves complex coordinate transformation calculations. For example, a homogeneous transformation matrix can be used to represent the position and orientation of the robot end effector:

[0158]

[0159] where R is a 3×3 rotation matrix representing the orientation, and p is a 3×1 translation vector representing the position. By multiplying the transformation matrices of each joint, the transformation matrix of the end effector relative to the base can be obtained:

[0160] T = T1·T2·...·T n ,

[0161] where n is the number of joints of the robot.

[0162] The position information in the standard coordinate system calculated by the coordinate transformation unit 14 will be sent to the data processing module 4 for subsequent trajectory analysis and optimization guidance.

[0163] Next, as Figure 3 shown, the optical tracker module 2 of the present invention includes an optical camera unit 21, a marker point recognition unit 22, an attitude calculation unit 23, and a data fusion unit 24. These units work together to achieve precise tracking of the learner and surgical instruments.

[0164] The optical camera unit 21 is used to collect image data of the surgical area. In a preferred embodiment of the present invention, the optical camera unit 21 includes at least three high-speed industrial cameras to form a stereo vision system. These cameras usually use global shutter CMOS sensors to avoid the rolling shutter effect when imaging moving objects. The resolution of the cameras can reach 2048×2048 pixels, and the frame rate can reach 120fps, so that fast hand movements can be captured without motion blur.

[0165] The marker point recognition unit 22 is communicatively connected to the optical camera unit 21 and is used to identify preset optical marker points from the image data. These marker points are usually attached to the learner's hand and surgical instruments. In an embodiment of the present invention, the marker points can be reflective spheres or LED light-emitting points. The marker point recognition unit 22 uses image processing algorithms, such as threshold segmentation and connected component analysis, to extract the two-dimensional coordinates of the marker points from the image. Then, through the corresponding point matching of multiple cameras, the three-dimensional spatial coordinates of the marker points are calculated. This process can be expressed as:

[0166] [X,Y,Z]=f(x1,y1,x2,y2,...,x n ,y n ),

[0167] where (x i ,y i ) are the two-dimensional coordinates of the marker point in the i-th camera, (X,Y,Z) are the three-dimensional spatial coordinates of the marker point, and f is the reconstruction function. The attitude calculation unit 23 is communicatively connected to the marker point recognition unit 22 and is used to calculate the attitude information of the learner and surgical instruments based on the spatial coordinates of the optical marker points. In a preferred embodiment of the present invention, the attitude calculation uses the least squares method to fit the rigid body transformation. Specifically, the following optimization problem can be used to solve the attitude:

[0168]

[0169] where R is the rotation matrix, t is the translation vector, pi and q i are the coordinates of the marker points in the local coordinate system and the global coordinate system respectively, and n is the number of marker points.

[0170] The data fusion unit 24 is communicatively connected to the attitude calculation unit 23 and is configured to fuse the spatial coordinates and attitude information to generate comprehensive spatial position information. In an embodiment of the present invention, the data fusion may adopt a Kalman filtering algorithm, taking the position and attitude information as the state vector to achieve smoothing and prediction. The state equation of the Kalman filter can be expressed as:

[0171] x k = Fx k-1 + w k ,

[0172] The observation equation is:

[0173] z k = Hx k + v k ,

[0174] where x k is the state vector, F is the state transition matrix, w k is the process noise, z k is the observation vector, H is the observation matrix, and v k is the observation noise.

[0175] The comprehensive spatial position information generated by the data fusion unit 24 will be sent to the data processing module 4 for subsequent trajectory analysis and optimization guidance.

[0176] Finally, as Figure 4 shown, the spatial force sensor module 3 of the present invention includes a force sensor array unit 31, a signal conditioning unit 32, a force vector calculation unit 33, and a pressure distribution mapping unit 34. These units work together to achieve precise measurement and analysis of the contact pressure of the surgical instrument.

[0177] The force sensor array unit 31 is mounted on the surgical instrument and is configured to collect multi-dimensional force information. In a preferred embodiment of the present invention, the force sensor array unit 31 is a six-axis force / torque sensor manufactured using MEMS (Micro-Electro-Mechanical System) technology. This sensor can simultaneously measure forces in three directions and torques in three directions, with a measurement range typically of ±100 N and ±5 Nm, and a resolution of up to 0.01 N and 0.001 Nm.

[0178] The signal conditioning unit 32 is electrically connected to the force sensor array unit 31 and is used to receive multi-dimensional force information and perform filtering and amplification processing on it. In an embodiment of the present invention, the signal conditioning unit 32 employs a multi-order Butterworth low-pass filter with a cut-off frequency set to 100 Hz to filter out high-frequency noise. At the same time, a programmable gain amplifier (PGA) is used to amplify the signal, and the gain range can be adjusted from 1 to 1000 times to adapt to different force ranges.

[0179] The force vector calculation unit 33 is communicatively connected to the signal conditioning unit 32 and is used to calculate the magnitude and direction of the force based on the processed multi-dimensional force information. In three-dimensional space, the force vector can be expressed as:

[0180]

[0181] where F x 、F y 、F z are the components of the force in the x, y, and z directions respectively, is the unit vector. The magnitude of the force can be calculated by the Euclidean norm:

[0182]

[0183] The direction of the force can be represented by two angles:

[0184]

[0185] where θ is the angle with the z-axis and φ is the angle between the projection on the xy plane and the x-axis.

[0186] The pressure distribution mapping unit 34 is communicatively connected to the force vector calculation unit 33 and is used to generate a pressure distribution map based on the magnitude and direction of the force. In a preferred embodiment of the present invention, the pressure distribution can be visualized in the form of a heatmap. Specifically, the contact surface of the surgical instrument can be discretized into an m×n grid, and the pressure value of each grid can be calculated by an interpolation algorithm. For example, using bilinear interpolation:

[0187] P(xy) = a0 + a1x + a2y + a3xy,

[0188] where P(x, y) is the pressure value at the point (x, y), and a0, a1, a2, and a3 are interpolation coefficients that can be solved from the known pressure values around.

[0189] The pressure distribution map generated by the pressure distribution mapping unit 34 will be sent to the data processing module 4 for subsequent trajectory analysis and optimization guidance. This visualized pressure distribution information is of great significance for evaluating the surgical skills of learners, especially their performance in controlling the applied force.

[0190] Through the close cooperation of the above-mentioned modules, the learner's motion trajectory analysis and optimization guidance system in surgical teaching of the present invention realizes the omnidirectional capture and analysis of surgical movements. The system can not only accurately track the hand movements of the learner, but also measure the contact pressure of surgical instruments, providing rich data support for the evaluation and guidance of surgical skills. This multi-modal data acquisition and analysis method is expected to significantly improve the surgical skill level and surgical safety of learners.

[0191] As Figure 5 shown, the data processing module 4 of the present invention includes a data synchronization unit 41, a noise filtering unit 42, a feature extraction unit 43, and a trajectory reconstruction unit 44. These units work together to realize the processing and fusion of multi-source data, laying a foundation for subsequent trajectory analysis.

[0192] The data synchronization unit 41 is used to receive the spatial position information from the optical tracker module 2 and the contact pressure data from the spatial force sensor module 3, and perform timestamp alignment on these data. In a preferred embodiment of the present invention, data synchronization is processed using the interpolation method. Specifically, linear interpolation or cubic spline interpolation algorithms can be used to align data with different sampling frequencies to the same time axis. For example, for linear interpolation, the following formula can be used:

[0193]

[0194] where (x1,y1) and (x2,y2) are known adjacent data points, and (x,y) is the data point to be interpolated. The noise filtering unit 42 is communicatively connected to the data synchronization unit 41 and is used to perform noise filtering processing on the synchronized data. In an embodiment of the present invention, noise filtering can be performed using the wavelet transform method. The wavelet transform has good time-frequency localization characteristics and can effectively remove high-frequency noise in the data while retaining the important features of the signal. For a one-dimensional signal f(t), its continuous wavelet transform can be expressed as:

[0195]

[0196] where ψ(t) is the wavelet basis function, a is the scale parameter, b is the translation parameter, and * represents the complex conjugate. By selecting an appropriate threshold for coefficient truncation, noise removal can be achieved.

[0197] The feature extraction unit 43 is communicatively connected to the noise filtering unit 42 and is used to extract key feature points from the processed data. In a preferred embodiment of the present invention, feature extraction is performed using the principal component analysis (PCA) method. PCA can project high-dimensional data into a low-dimensional space and retain the main features of the data. Specifically, for the data matrix X, its covariance matrix is:

[0198]

[0199] By solving the characteristic equation:

[0200] Cv = λv,

[0201] the eigenvalues λ and the corresponding eigenvectors v can be obtained. Selecting the eigenvectors corresponding to the largest several eigenvalues can obtain the principal components.

[0202] The trajectory reconstruction unit 44 is communicatively connected to the feature extraction unit 43 and is used to reconstruct the action trajectory of the learner based on the key feature points. In an embodiment of the present invention, spline interpolation method can be used for trajectory reconstruction. For example, for cubic spline interpolation, a piecewise cubic polynomial function can be constructed as follows:

[0203] S i (x) = a i + b i (x - x i ) + c i (x - x i ) 2 + d i (x - x i ) 3 , x ∈ [x i , x i+1 ,

[0204] where a i , b i , c i , d i are undetermined coefficients and can be obtained by solving a system of linear equations.

[0205] The learner action trajectory data generated by the trajectory reconstruction unit 44 will be sent to the trajectory analysis module 5 for subsequent analysis and evaluation.

[0206] As Figure 6 shown, the trajectory analysis module 5 of the present invention includes a standard trajectory storage unit 51, a trajectory comparison unit 52, a deviation calculation unit 53, and a comprehensive evaluation unit 54. These units work together to achieve quantitative analysis and evaluation of the learner's action trajectory.

[0207] The standard trajectory storage unit 51 is used to store preset standard action trajectory data. In a preferred embodiment of the present invention, the standard trajectory data is represented by a B-spline curve. The B-spline curve has good local controllability and smoothness and is suitable for representing complex surgical action trajectories. The nth-order B-spline curve can be represented as:

[0208]

[0209] Among them, P i is the control point, and N i,k (t) is the k-th order B-spline basis function. The trajectory comparison unit 52 is communicatively connected to the standard trajectory storage unit 51, and is configured to receive the learner's action trajectory data and compare it with the standard action trajectory data. In an embodiment of the present invention, the trajectory comparison adopts the dynamic time warping (DTW) algorithm. The DTW algorithm can process time series of different lengths and speeds and calculate the similarity between them. For two time series A = (a1,..., a n ) and B = (b1,..., b m ), the DTVV distance can be solved by dynamic programming:

[0210]

[0211] Among them, d(a i , b j ) is the distance between point a i and b j .

[0212] The deviation calculation unit 53 is communicatively connected to the trajectory comparison unit 52, and is configured to calculate the spatial deviation and the time deviation between the learner's action trajectory and the standard action trajectory. The spatial deviation can be represented by the Euclidean distance:

[0213]

[0214] Among them, (x i , y i , z i ) is the point on the learner's trajectory, and x′ i , y′ i , z′ i ) is the corresponding point on the standard trajectory. The time deviation can be represented by the speed difference:

[0215]

[0216] Among them, v i and v′ i are the speeds of the corresponding points on the learner's trajectory and the standard trajectory respectively.

[0217] The comprehensive evaluation unit 54 is communicatively connected to the deviation calculation unit 53, and is configured to calculate a comprehensive deviation index based on the spatial deviation and the time deviation. In a preferred embodiment of the present invention, the comprehensive deviation index adopts the form of a weighted sum:

[0218] D = αD s + (1 - α)D t ,

[0219] Among them, α is a weight coefficient (0 ≤ α ≤ 1), which can be adjusted according to the requirements of different surgical types. For example, for fine operations, the weight of spatial deviation can be increased; for operations that require quick response, the weight of time deviation can be increased.

[0220] The comprehensive deviation index calculated by the comprehensive evaluation unit 54 will be sent to the optimization guidance module 6 for generating targeted optimization suggestions.

[0221] As Figure 7 shown, the optimization guidance module 6 of the present invention includes an expert knowledge base unit 61, a rule matching unit 62, a suggestion generation unit 63, and a priority ranking unit 64. These units work together to achieve intelligent guidance based on expert experience and data analysis.

[0222] The expert knowledge base unit 61 is used to store preset optimization rules for surgical actions. In a preferred embodiment of the present invention, the knowledge base is organized in an ontology structure. An ontology can be represented by a triple (S, P, O), where S is the subject, P is the predicate, and O is the object. For example, an optimization rule can be represented in the following form:

[0223] (Fast_Movement, causes, Accuracy_Decrease)

[0224] (Accuracy_Decrease, suggest, Slow_Down).

[0225] The rule matching unit 62 is communicatively connected to the expert knowledge base unit 61 and is used to receive the trajectory deviation index and match the corresponding optimization rule from the expert knowledge base according to this index. In an embodiment of the present invention, rule matching can adopt a fuzzy inference system. For example, using the Mamdani inference model, the deviation index can be used as the input, and through the processes of fuzzification, rule inference, and defuzzification, the matching optimization rule can be obtained.

[0226] The suggestion generation unit 63 is communicatively connected to the rule matching unit 62 and is used to generate specific optimization suggestions based on the matched optimization rules. The optimization suggestions usually include the direction and amplitude of action adjustment. For example, for spatial deviation, a suggestion in the following form can be generated: Please move the surgical instrument in the {direction} by approximately {distance} mm. Among them, the direction and distance can be calculated by analyzing the trajectory deviation data.

[0227] The priority ranking unit 64 is communicatively connected to the suggestion generation unit 63 and is used to rank the generated optimization suggestions in terms of priority. In a preferred embodiment of the present invention, multi-criteria decision-making methods are adopted for priority ranking. For example, the analytic hierarchy process (AHP) can be used to construct a decision-making model. The priority can be expressed as:

[0228]

[0229] Among them, P i is the priority of the i-th suggestion, w j is the weight of the j-th criterion, and c ij is the score of the i-th suggestion under the j-th criterion.

[0230] The priority sorting unit 64 sends the sorted optimized suggestions to the feedback execution module 7 for actual feedback guidance.

[0231] Through the collaborative work of the above modules, the learner's motion trajectory analysis and optimization guidance system of the present invention realizes the full-process intelligence from data processing, trajectory analysis to optimization guidance. The system can not only accurately capture and analyze the surgical actions of learners, but also generate targeted optimization suggestions based on expert knowledge and data analysis. Such an intelligent teaching assistance system provides a strong tool support for improving the surgical teaching effect and the skill level of learners.

[0232] The system of the present invention further includes a learner model construction module 8, which is communicatively connected to the data processing module 4 and the trajectory analysis module 5. The introduction of the learner model construction module 8 enables the system to provide more personalized teaching guidance according to the individual characteristics and learning history of learners.

[0233] The learner model construction module 8 is mainly used to receive the historical motion trajectory data and trajectory deviation indicators of learners, and construct a personalized skill model of learners based on these historical data. In the preferred embodiment of the present invention, the personalized skill model is represented by a Hidden Markov Model (HMM). HMM is a statistical model, which is particularly suitable for describing processes with time series characteristics, such as surgical action sequences.

[0234] HMM can be represented by a five-tuple λ = (S, V, A, B, π), where:

[0235] S = {s1, s2,..., s N} is the set of hidden states, representing different surgical skill stages.

[0236] V = {v1, v2,..., v M} is the set of observation symbols, representing observable surgical action characteristics.

[0237] A = {a ij} is the state transition probability matrix, a ij = P(q t+1 = s j |q t = si )

[0238] B = {b j (k)} is the observation probability matrix, where b j (k) = P(o t = v k |q t = s j )

[0239] π = {π i} is the initial state distribution, and π i = P(q1 = s i )

[0240] The learner model construction module 8 trains the parameters of the HMM through the Baum-Welch algorithm. Given the observation sequence O = (o1, o2,..., o T ), the goal of the algorithm is to find the optimal model parameter λ * , such that:

[0241]

[0242] After training, the learner model construction module 8 sends the constructed personalized skill model to the optimization guidance module 6 for generating targeted optimization suggestions. For example, the system can compare the learner's current action sequence with the optimal sequence predicted by the model to find the key points that need improvement, thereby providing more accurate guidance.

[0243] After training, the learner model construction module 8 sends the constructed personalized skill model to the optimization guidance module 6 for generating targeted optimization suggestions. For example, the system can compare the learner's current action sequence with the optimal sequence predicted by the model to find the key points that need improvement, thereby providing more accurate guidance.

[0244] In addition, the learner model construction module 8 also has the function of adaptive learning. As the learner's skill level improves, the module will continuously update the model parameters to ensure that the guidance suggestions always match the learner's actual level. This dynamic adjustment mechanism greatly improves the adaptability and effectiveness of the system.

[0245] Finally, the system of the present invention further includes a virtual reality interaction module 9, which is communicatively connected to the hand-eye coordination robot module 1 and the data processing module 4. The introduction of the virtual reality interaction module 9 provides an immersive learning environment for the learner, greatly enhancing the teaching effect and user experience of the system.

[0246] The virtual reality interaction module 9 is mainly used to generate a virtual surgical environment, in which the movement trajectory of the learner is displayed in real time, and haptic feedback of virtual surgical instruments is provided. In a preferred embodiment of the present invention, the virtual reality environment is constructed by combining three-dimensional computer graphics technology and a physics engine.

[0247] Specifically, the rendering of the virtual surgical environment can be implemented using graphics APIs such as OpenGL or DirectX. To improve the rendering efficiency and realism, the system adopts the level of detail (LOD) technology and the physically based rendering (PBR) method. The LOD technology can dynamically adjust the complexity of the model according to the distance of the object from the viewpoint, and its mathematical model can be expressed as:

[0248]

[0249] where L(d) is the level of detail at distance d, L0 is the highest level of detail, d max is the maximum visible distance, and n is the attenuation coefficient. The PBR method realizes a more realistic material effect by simulating the physical behavior of light. Its core is the bidirectional reflectance distribution function (BRDF) based on the microfacet theory:

[0250]

[0251] where D is the normal distribution function, F is the Fresnel term, G is the geometric shadowing function, ω i 、ω o and ω h are the incident light direction, the outgoing light direction, and the half-way vector, respectively. The virtual reality interaction module 9 is also responsible for providing haptic feedback, which is crucial for simulating the real surgical feeling. Haptic feedback is usually realized through a force feedback device, and its control algorithm can be based on a virtual coupling model:

[0252]

[0253] where F is the output force, k p and k d are the stiffness coefficients of position and velocity, x v and x r are the positions of the virtual object and the real device, respectively, and are the corresponding velocities.

[0254] The virtual reality interaction module 9 not only provides visual and haptic feedback, but also can record the operation data of the learner in the virtual environment. These data will be sent to the data processing module 4 for fusion analysis with the data collected in the real environment. This combination of virtual and real enables the system to comprehensively evaluate the performance of the learner and provide more comprehensive teaching guidance.

[0255] By introducing the learner model construction module 8 and the virtual reality interaction module 9, the learner motion trajectory analysis and optimization guidance system in the surgical teaching of the present invention realizes a personalized, intelligent and immersive teaching experience. The system can not only provide customized guidance according to the individual characteristics of learners, but also create a nearly real surgical environment through virtual reality technology, greatly improving the learning effect and teaching efficiency. This innovative teaching system combining artificial intelligence, virtual reality and robot technology is expected to significantly improve the quality and efficiency of medical talent cultivation.

[0256] To verify the superiority of the learner motion trajectory analysis and optimization guidance system in the surgical teaching of the present invention, a group of simulation experiments were designed to simulate the teaching scenario of laparoscopic cholecystectomy. 30 surgical interns were selected as test subjects, randomly divided into three groups of 10 people each, and trained for 4 weeks using different teaching methods.

[0257] Example: Train using the surgical teaching system of the present invention.

[0258] Comparative Example 1: Use traditional simulation training under the guidance of a tutor.

[0259] Comparative Example 2: Use an existing virtual reality surgical simulator for training, but without the motion trajectory analysis and optimization guidance function of the present invention.

[0260] The experimental conditions are set as follows:

[0261] 1. Training duration: Each trainee conducts 3 training sessions per week, and each training session lasts for 2 hours.

[0262] 2. Evaluation task: Before and after the training, all trainees need to complete a standardized laparoscopic cholecystectomy simulation task.

[0263] 3. Evaluation indicators: We selected the following key indicators to evaluate the performance of trainees:

[0264] a) Operation completion time (minutes);

[0265] b) Motion trajectory accuracy (mm, average deviation from the standard trajectory);

[0266] c) Instrument operation force control (N, average deviation from the ideal force);

[0267] d) Correct rate of key steps (%);

[0268] e) Incidence rate of complications (%);

[0269] f) Trainee subjective satisfaction (1 - 10 points);

[0270] Evaluation method:

[0271] 1. Operation completion time: Automatically recorded by the system.

[0272] 2. Movement trajectory accuracy: Record the movement trajectory of the trainee's instrument through the optical tracking system, compare it with the preset expert standard trajectory, and calculate the average deviation.

[0273] 3. Instrument operation force control: Record the force application during the trainee's operation through the force sensor, compare it with the preset ideal force range, and calculate the average deviation.

[0274] 4. Correct rate of key steps: Individually scored by three senior surgeons based on the video, and the average value is taken.

[0275] 5. Incidence of complications: Set potential complication triggering conditions in the simulation environment and record the number of triggers.

[0276] 6. Trainee's subjective satisfaction: Obtained through anonymous questionnaires.

[0277] The experimental results are shown in the following table:

[0278]

[0279]

[0280] Analysis and discussion:

[0281] 1. Operation completion time: The system of the present invention is most prominent in shortening the operation time, improving by 14.2% and 7.4% respectively compared with the traditional method and the ordinary virtual reality system. This is mainly due to the real-time motion analysis and optimization guidance of the system, which helps trainees master efficient operation skills faster.

[0282] 2. Movement trajectory accuracy: The system of the present invention has the most significant effect in improving movement accuracy, being 1.4 mm and 0.7 mm higher than the traditional method and the ordinary virtual reality system respectively. This benefits from the high-precision optical tracking and real-time trajectory analysis functions of the system, which can correct the movement deviation of trainees in a timely manner.

[0283] 3. Force control: The system of the present invention also performs well in improving force control, being 0.9 N and 0.5 N higher than the other two methods respectively. This is mainly attributed to the integrated force sensor and force feedback device of the system, which provide precise tactile feedback for trainees.

[0284] 4. Correct rate of key steps: The system of the present invention is most effective in improving the correct rate of key steps, being 9.4% and 5.2% higher than the traditional method and the ordinary virtual reality system respectively. This reflects that the intelligent guidance function of the system can effectively help trainees master the key skills of surgery.

[0285] 5. Incidence rate of complications: The system of the present invention is most prominent in reducing the incidence rate of complications, being 26.2% and 14.6% higher than the other two methods respectively. This indicates that the real-time monitoring and early warning functions of the system can effectively prevent potential surgical risks.

[0286] 6. Trainee satisfaction: The system of the present invention has obtained the highest trainee satisfaction score, reflecting that its personalized and intelligent teaching method has been recognized by trainees.

[0287] In summary, the surgical teaching system of the present invention is significantly superior to the traditional method and the ordinary virtual reality system in all key indicators. This advantage mainly stems from the system integrating high-precision motion capture technology, intelligent trajectory analysis algorithms, personalized learning models, and immersive virtual reality environments. Especially in improving motion accuracy and reducing the risk of complications, the system of the present invention shows significant advantages, which is of great significance for improving surgical safety and patient prognosis.

[0288] In addition, another significant advantage of the system of the present invention is its adaptability and scalability. By adjusting parameters and updating the knowledge base, the system can easily adapt to different types of surgical training needs.

[0289] It should be noted that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A learner motion trajectory analysis and optimization guidance system in surgical teaching, characterized in that, Including: A hand-eye coordinated robot module, configured to: Execute simulated surgical actions according to preset instructions; Collect the position and attitude information of the end effector of the robot; An optical tracker module, communicatively connected to the hand-eye coordinated robot module, configured to: Realtime capture the spatial position information of the learner and the surgical instrument; Send the spatial position information to the data processing module; A spatial force sensor module, physically connected to the hand-eye coordinated robot module, configured to: Detect the contact pressure of the surgical instrument; Transmit the contact pressure data to the data processing module; A data processing module, communicatively connected to the optical tracker module and the spatial force sensor module, configured to: Receive the spatial position information and the contact pressure data; Generate learner action trajectory data based on the spatial position information and the contact pressure data; A trajectory analysis module, communicatively connected to the data processing module, configured to: Receive the learner action trajectory data; Compare the learner action trajectory data with a preset standard action trajectory; Calculate a trajectory deviation index based on the comparison result; An optimization guidance module, communicatively connected to the trajectory analysis module, configured to: Receive the trajectory deviation index; Generate an optimization suggestion according to the trajectory deviation index; Send the optimization suggestion to the feedback execution module; A feedback execution module, communicatively connected to the optimization guidance module and the hand-eye coordinated robot module, configured to: Receive the optimization suggestion; Control the hand-eye coordinated robot module to execute corresponding feedback actions.

2. The system according to claim 1, characterized in that The hand-eye coordinated robot module includes: A robot body unit, configured to execute preset surgical actions; A robotic arm base unit, physically connected to the robot body unit, configured to support and fix the robot body unit; A spatial position sensor unit, disposed on the robotic arm base unit, configured to: Collect the spatial position data of the end effector of the robot body unit; Send the spatial position data to the coordinate conversion unit; A coordinate conversion unit, communicatively connected to the spatial position sensor unit, configured to: Receive the spatial position data; Convert the spatial position data into position information in a standard coordinate system; Send the position information in the standard coordinate system to the data processing module.

3. The system according to claim 1, wherein The optical tracker module includes: An optical camera unit, configured to collect image data of the surgical area; A marker point recognition unit, communicatively connected to the optical camera unit, configured to: Identify preset optical marker points from the image data; Calculate the spatial coordinates of the optical marker points; An attitude calculation unit, communicatively connected to the marker point recognition unit, configured to: Calculate the attitude information of the learner and the surgical instrument based on the spatial coordinates of the optical marker points; A data fusion unit, communicatively connected to the attitude calculation unit, configured to: Fuse the spatial coordinates and the attitude information; Generate comprehensive spatial position information; Send the comprehensive spatial position information to the data processing module.

4. The system according to claim 1, wherein The spatial force sensor module includes: A force sensor array unit, installed on the surgical instrument, configured to collect multi-dimensional force information; A signal conditioning unit, electrically connected to the force sensor array unit, for: Receiving the multi-dimensional force information; Filtering and amplifying the multi-dimensional force information; A force vector calculation unit, communicatively connected to the signal conditioning unit, for: Calculating the magnitude and direction of the force based on the processed multi-dimensional force information; A pressure distribution mapping unit, communicatively connected to the force vector calculation unit, for: Generating a pressure distribution map according to the magnitude and direction of the force; Sending the pressure distribution map to the data processing module.

5. The system according to claim 1, wherein The data processing module includes: A data synchronization unit, for: Receiving the spatial position information from the optical tracker module and the contact pressure data from the spatial force sensor module; performing timestamp alignment on the spatial position information and the contact pressure data; A noise filtering unit, communicatively connected to the data synchronization unit, for: Performing noise filtering on the synchronized data; A feature extraction unit, communicatively connected to the noise filtering unit, for: Extracting key feature points from the processed data; A trajectory reconstruction unit, communicatively connected to the feature extraction unit, for: Reconstructing the action trajectory of the learner based on the key feature points; Generating learner action trajectory data; Sending the learner action trajectory data to the trajectory analysis module.

6. The system according to claim 1, wherein The trajectory analysis module includes: a standard trajectory storage unit for storing preset standard action trajectory data; A trajectory comparison unit, communicatively connected to the standard trajectory storage unit, for: Receiving learner action trajectory data; Comparing the learner action trajectory data with the standard action trajectory data; A deviation calculation unit, communicatively connected to the trajectory comparison unit, for: Calculating the spatial deviation between the learner action trajectory and the standard action trajectory; Calculating the time deviation between the learner action trajectory and the standard action trajectory; A comprehensive evaluation unit, communicatively connected to the deviation calculation unit, for: Calculating a comprehensive deviation index based on the spatial deviation and the time deviation; Sending the comprehensive deviation index to the optimization guidance module.

7. The system according to claim 1, characterized in that, The optimization guidance module includes: an expert knowledge base unit for storing preset surgical action optimization rules; A rule matching unit, communicatively connected to the expert knowledge base unit, for: Receiving the trajectory deviation index; Matching corresponding optimization rules from the expert knowledge base according to the trajectory deviation index; A suggestion generation unit, communicatively connected to the rule matching unit, for: Generating specific optimization suggestions based on the matched optimization rules; A priority ranking unit, communicatively connected to the suggestion generation unit, for: Performing priority ranking on the generated optimization suggestions; Sending the ranked optimization suggestions to the feedback execution module.

8. The system according to claim 1, characterized in that, The feedback execution module includes: A visual feedback unit for presenting the visualization information of the optimization suggestions on a display device; A tactile feedback unit, communicatively connected to the hand-eye coordination robot module, for: Controlling the hand-eye coordination robot module to generate simulated force feedback; A voice feedback unit for converting the optimization suggestions into voice prompts; A feedback coordination unit, communicatively connected to the visual feedback unit, the tactile feedback unit, and the voice feedback unit, is configured to: Coordinate the execution order and intensity of different feedback methods according to the content and priority of the optimization suggestions.

9. The system according to claim 1, characterized in that It further includes: A learner model construction module, communicatively connected to the data processing module and the trajectory analysis module, is configured to: Receive the historical action trajectory data and trajectory deviation metrics of the learner; Construct a personalized skill model of the learner based on the historical data; Send the personalized skill model to the optimization guidance module for generating targeted optimization suggestions.

10. The system according to claim 1, wherein It further includes: A virtual reality interaction module, communicatively connected to the hand-eye coordination robot module and the data processing module, is configured to: Generate a virtual surgical environment; Realtime display the action trajectory of the learner in the virtual surgical environment; Provide tactile feedback of virtual surgical instruments; Wherein, the virtual reality interaction module is further configured to send the interaction data in the virtual environment to the data processing module to achieve data fusion analysis of the real environment and the virtual environment.

Images