Bronchoscope operation training method and system for endoscopic surgery

Abstract

The invention provides a bronchoscope operation training method and system for endoscopic surgery, and relates to the technical field of virtual training, and the method comprises the steps: recognizing the tissue springback time when the lens advancing speed exceeds an airway position of a preset safety interval, extracting a tissue injury high-risk section with a tissue springback time deviation value exceeding a standard range, and obtaining a tissue injury high-risk section; the high-risk section of tissue injury is visualized; extracting the time fluctuation amplitude of tissue compression and recovery of the tissue damage high-risk section, and evaluating the difference between the time fluctuation amplitude and a preset deformation response duration upper limit to obtain an initial position record of a tissue damage risk point; and according to an upper limit threshold of the propulsion speed of the airway section position, evaluating a deviation value with the current lens propulsion speed, generating a speed reduction amplitude when the upper limit threshold is lower than the current speed, obtaining a real-time speed limit threshold of each airway section, and displaying the real-time speed limit threshold on a training interface in real time. And through multi-dimensional data fusion and a real-time feedback mechanism, the skills of the trainees are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of virtual training, and more particularly to a method and system for training bronchoscopic procedures in laparoscopic surgery. Background Technology

[0002] In the field of medical training, bronchoscopic surgery training for laparoscopic procedures holds a crucial position, directly impacting the skill level of medical staff and patient safety. By simulating a real surgical environment, this training helps trainees master complex airway examination and manipulation techniques, offering irreplaceable value in enhancing clinical practice. However, despite the high level of attention given to this area, existing training methods in practical application primarily rely on the operator's subjective judgment and accumulated experience, lacking real-time monitoring and feedback mechanisms for dynamic changes during the procedure. This leads to poor training effectiveness and may even increase operational risks. Particularly when facing different trainees' operating habits and the tissue characteristics of different airway sites, the system cannot provide personalized guidance, easily causing confusion and hindering accurate skill mastery. Operating speed, as a key factor, directly affects training efficiency and operational safety. If the speed is too fast, the elastic deformation of the airway wall cannot be reflected in the operator's tactile feedback in time, making it difficult for trainees to accurately perceive the true state of the tissue; conversely, if the speed is too slow, it significantly prolongs the operation time and reduces training efficiency. This contradiction is particularly prominent in actual operation. For example, during simulated airway examination, trainees, in pursuit of speed, advance the lens quickly, but because they cannot promptly perceive subtle changes in the airway walls, they repeatedly adjust the position, wasting time and potentially damaging the simulated tissue due to excessive force. This phenomenon of lack of feedback and operational errors caused by improper speed control has become a core obstacle that urgently needs to be addressed in training. Therefore, how to reasonably control the lens advancement speed in bronchoscopy operation training while ensuring timely and accurate tissue feedback has become a key issue in improving the safety and efficiency of training. Summary of the Invention

[0003] This invention provides a bronchoscopic operation training method for laparoscopic surgery, comprising: A training simulation model is constructed, in which a speed monitoring module and a tissue deformation sensing module are installed. This model acquires the lens advance speed, airway tissue rebound time, and handle feedback force intensity based on the trainee's operation. It identifies airway locations where the lens advance speed exceeds a preset safe range; identifies the tissue rebound time at airway locations where the lens advance speed exceeds the preset safe range; extracts high-risk tissue damage sections where the tissue rebound time deviation exceeds the standard range; and visualizes these high-risk tissue damage sections on the training interface. It also extracts the time fluctuation amplitude of tissue compression and recovery in these high-risk tissue damage sections, assesses the difference between this time fluctuation amplitude and a preset upper limit for deformation response time, and obtains a preliminary location record of the tissue damage risk point. Finally, it analyzes the handle feedback force intensity corresponding to the preliminary location record of the tissue damage risk point and compares it with the training simulation model. The system compares the standard tactile feedback intensity benchmarks for each airway segment to identify risk points where the feedback intensity is lower than the standard benchmark. It then assesses the maximum allowable propulsion speed for the corresponding airway segment position, obtains the upper limit threshold for propulsion speed of the airway segment position, and feeds it back to the trainee. Based on the upper limit threshold for propulsion speed of the airway segment position, it assesses the deviation from the current camera propulsion speed. When the upper limit threshold is lower than the current speed, it generates a deceleration rate, obtains the real-time speed limit threshold for each airway segment, and displays it in real time on the training interface. The system then compares the trainee's actual propulsion speed in each airway segment with the real-time speed limit threshold for each airway segment to determine the airway segment positions that require speed control. It identifies the distribution of airway segments where trainees frequently violate operating procedures, analyzes the trainee's operational weaknesses in speed control, tissue recognition, and force perception, and generates targeted suggestions for improving operational skills.

[0004] Furthermore, the construction of the training simulation model, in which a speed monitoring module and a tissue deformation sensing module are installed, acquires the lens advance speed, airway segment tissue rebound time, and handle end feedback force intensity of the trainee's operation, and identifies airway locations where the lens advance speed exceeds the preset safe range. This includes: acquiring three-dimensional airway structure data of the bronchoscopy training simulation model; deploying fiber optic sensor arrays at the bifurcation locations of the main bronchus, lobar bronchus, and segmental bronchus; recording the advance speed value and corresponding timestamp when the lens passes through each airway segment; calculating the instantaneous speed based on the trigger time difference of adjacent sensors; recording the airway segment position number when the speed exceeds the upper limit of the safe range; deploying a piezoelectric sensor at the airway segment position at the moment of speed exceeding the limit; collecting the deformation of the airway wall after pressure; recording the time interval for the tissue to recover to its original shape after pressure release as the tissue rebound time; and using the strain gauge sensor built into the force feedback handle to measure the handle end feedback force intensity value.

[0005] Furthermore, the process involves identifying the tissue rebound time at airway locations where the lens advance speed exceeds a preset safety range, extracting high-risk tissue damage segments where the tissue rebound time deviation exceeds the standard range, and visualizing these high-risk segments on the training interface. This includes: acquiring airway location data where the lens advance speed exceeds the preset safety range, reading the corresponding tissue rebound time measurement value, calculating the difference with a preset standard rebound time benchmark value to obtain a risk point dataset containing location coordinates and rebound time deviation values, grouping the risk points using a K-means clustering algorithm, using the deviation value as the primary feature and the three-dimensional coordinate distance as the auxiliary feature, aggregating risk points with similar deviation values ​​and continuous spatial locations into high-risk tissue damage segments, calculating the average deviation value and spatial range of each high-risk segment to obtain a boundary coordinate sequence of the high-risk segment, rendering the region on a three-dimensional airway model using the boundary coordinate sequence, assigning different color intensities according to the magnitude of the average deviation value, and overlaying and displaying the results on the training interface.

[0006] Furthermore, the temporal fluctuation amplitude of tissue compression and recovery in the high-risk tissue damage area is extracted, and the difference between the temporal fluctuation amplitude and the preset upper limit of deformation response duration is evaluated to obtain a preliminary location record of the tissue damage risk point. This includes: extracting the pressure application time and the tissue morphology recovery time after pressure release from each sampling point in the high-risk tissue damage area; calculating the duration of the compression phase and the duration of the recovery phase; recording the time data of multiple consecutive compression and recovery cycles; calculating the total duration of each cycle and obtaining the standard deviation to obtain the temporal fluctuation amplitude value; comparing the temporal fluctuation amplitude value with the preset upper limit of deformation response duration; marking sampling points whose fluctuation amplitude exceeds the upper limit value by a preset percentage threshold as potential risk points; and recording the three-dimensional coordinates and the degree of fluctuation amplitude exceeding the limit of the potential risk points.

[0007] Furthermore, the initial location records of the tissue damage risk points are analyzed, and the corresponding handpiece feedback force intensity is compared with the standard tactile feedback intensity benchmark for each airway segment in the training simulation model. Risk points with feedback force lower than the standard benchmark are identified, and the maximum allowable propulsion speed of the airway segment corresponding to the risk point is evaluated. The upper limit threshold of propulsion speed for the airway segment is obtained and fed back to the trainee. This includes: obtaining the initial location records of the tissue damage risk points; reading the trainee's handpiece feedback force intensity value corresponding to each risk point location; calculating the difference between the measured feedback force and the standard benchmark; marking points with negative differences and absolute values ​​exceeding a preset percentage threshold of the benchmark value as insufficient feedback risk points; reading the preset tissue attribute parameters of the corresponding airway segment from the training simulation model; establishing a mapping relationship using a linear regression method with the force deviation as the independent variable and the safe propulsion speed in the historical operation record as the dependent variable; calculating the maximum allowable propulsion speed value of the airway segment where each insufficient feedback risk point is located through the mapping relationship; taking the minimum maximum allowable propulsion speed of multiple risk points within the same airway segment and multiplying it by a preset safety factor to obtain the upper limit threshold of propulsion speed for the airway segment.

[0008] Furthermore, based on the upper limit threshold of the propulsion speed at the location of the airway segment, the deviation from the current lens propulsion speed is evaluated. When the upper limit threshold is lower than the current speed, a reduction rate is generated, and the real-time speed limit thresholds for each airway segment are obtained and displayed in real time on the training interface. This includes: obtaining the upper limit threshold of the propulsion speed at the location of the airway segment and the current lens propulsion speed value; calculating the difference between the current speed and the upper limit threshold; determining the reduction rate based on the overspeed percentage; combining the difference in the upper limit threshold of the speed between adjacent airway segments; smoothing the reduction rate using a moving average method; obtaining a real-time speed limit threshold sequence for each airway segment; constructing a dynamic speed display layer on the training interface; and using color coding to identify different speed limit levels.

[0009] Furthermore, the actual propulsion speed of trainees in each airway segment is compared with the real-time speed limit threshold of each airway segment to determine the location of airway segments requiring speed control, identify the distribution of airway segments with high-frequency violations by trainees, analyze the trainees' operational weaknesses in speed control, tissue recognition, and force perception, and generate targeted operational skill improvement suggestions. This includes: extracting the trainees' actual propulsion speed sequence in each airway segment from the training record database, calculating the speed exceedance value at each sampling point, statistically analyzing the violation rate distribution data of each airway segment, screening airway segments with violation rates exceeding a preset threshold as high-frequency violation segments, using the K-means clustering algorithm to classify high-frequency violation segments with similar characteristics, extracting the corresponding speed control deviation value, tissue deformation response time deviation, and feedback force change rate data, and identifying the dimension with the greatest difference as the main weakness.

[0010] On the other hand, the present invention also discloses a bronchoscopic operation training system for laparoscopic surgery, the system comprising: The speed monitoring and tissue deformation sensing module is used to construct a training simulation model. The speed monitoring module and the tissue deformation sensing module are installed in the training simulation model to obtain the lens advance speed, airway tissue rebound time and handle end feedback force intensity of the trainee's operation, and to identify the airway position where the lens advance speed exceeds the preset safe range. The high-risk tissue damage segment identification and visualization module is used to identify the tissue rebound time at the airway location where the lens advance speed exceeds the preset safety range, extract the high-risk tissue damage segment where the tissue rebound time deviation value exceeds the standard range, and visualize the high-risk tissue damage segment on the training interface. The preliminary location module for tissue damage risk points is used to extract the time fluctuation amplitude of tissue compression and recovery in the high-risk area of ​​tissue damage, evaluate the difference between the time fluctuation amplitude and the preset upper limit of deformation response time, and obtain a preliminary location record of the tissue damage risk point. The feedback intensity analysis and propulsion speed upper limit assessment module is used to analyze the initial location record of the tissue damage risk point and the corresponding handpiece end feedback intensity, compare it with the standard tactile feedback intensity benchmark of each airway segment in the training simulation model, identify risk points where the feedback intensity is lower than the standard benchmark, assess the maximum bearing propulsion speed of the airway segment corresponding to the risk point, obtain the propulsion speed upper limit threshold of the airway segment position and provide feedback to the trainee. The real-time speed limit threshold generation and display module is used to evaluate the deviation from the current lens advance speed based on the upper limit threshold of the advance speed of the airway segment position, generate the deceleration range when the upper limit threshold is lower than the current speed, obtain the real-time speed limit threshold of each airway segment, and display it in real time on the training interface. The module for analyzing operational weaknesses and generating skill improvement suggestions is used to statistically analyze the actual propulsion speed of trainees in each airway segment, compare it with the real-time speed limit threshold of each airway segment, determine the location of airway segments that require speed control, identify the distribution of airway segments where trainees frequently violate operating rules, analyze the trainees' operational weaknesses in speed control, tissue recognition, and force perception, and generate targeted operational skill improvement suggestions.

[0011] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: This invention discloses a training method for bronchoscopy operation in laparoscopic surgery. Addressing the challenges faced by trainees in airway manipulation, including exceeding lens advance speed limits, failing to identify tissue damage risks, and lacking sufficient control over operational force, the method constructs a simulation model integrating speed monitoring and tissue deformation sensing modules. This model acquires trainee operational data in real time, identifies airway locations where speed exceeds limits and tissue rebound time deviations, extracts and visualizes high-risk sections, and analyzes time fluctuations and deformation response discrepancies to initially pinpoint risk points. Subsequently, by comparing the handle feedback force with standard benchmarks, the method determines the upper limit threshold for advance speed corresponding to the risk points and generates a speed limit threshold displayed in real time on the interface. Simultaneously, it analyzes trainees' actual operational data, identifies frequently violated procedures, analyzes operational weaknesses, and provides targeted suggestions. This invention, through multi-dimensional data fusion and real-time feedback mechanisms, significantly improves trainees' skills in speed control, tissue identification, and force perception, ensuring operational safety and accuracy. Attached Figure Description

[0012] Figure 1 This is a flowchart of a bronchoscopic operation training method for laparoscopic surgery according to the present invention.

[0013] Figure 2 This is a schematic diagram of a bronchoscopic operation training method for laparoscopic surgery according to the present invention.

[0014] Figure 3 This is a schematic diagram of a bronchoscopic operation training system for laparoscopic surgery according to the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and thoroughly described below with reference to the accompanying drawings. The described embodiments are merely some embodiments of the present invention.

[0016] like Figures 1 to 3 This embodiment of a bronchoscopic operation training method and system for laparoscopic surgery may specifically include: Step S101: Construct a training simulation model. By installing a speed monitoring module and a tissue deformation sensing module in the training simulation model, the camera advance speed, airway tissue rebound time, and handle end feedback force intensity of the trainee's operation are obtained, and the airway position where the camera advance speed exceeds the preset safe range during the trainee's operation is identified.

[0017] Three-dimensional airway structure data from a bronchoscopy training simulation model is acquired. Fiber optic sensor arrays are deployed at the bifurcation points of the main bronchus, lobar bronchus, and segmental bronchus to record the lens advance velocity and corresponding timestamps as it passes through each airway segment. Instantaneous velocity is calculated based on the trigger time difference between adjacent sensors, resulting in a lens advance velocity sequence for each airway segment. This sequence is compared with a preset safe velocity range for each airway segment. When the advance velocity of an airway segment exceeds the upper limit of the safe range, the segment's location number and the moment of velocity exceedance are recorded. A piezoelectric sensor is deployed at the airway segment location at the moment of velocity exceedance to collect the deformation of the airway wall after compression. The time interval between pressure release and tissue recovery to its original shape is recorded as the tissue rebound time. When the tissue rebound time exceeds a preset standard rebound time threshold for that airway segment, it is marked as a key monitoring area for velocity control. The strain gauge sensor built into the force feedback handle measures the operating reaction force in the key monitoring area of ​​the speed control, and obtains the intensity value of the feedback force generated by the contact between the lens and the airway wall. When the intensity of the feedback force is lower than the lower limit of the preset force safety range and the duration exceeds the preset duration threshold, the airway segment is determined to be the airway position where the lens advance speed exceeds the preset safety range.

[0018] For example, in one implementation, the bronchoscopy training simulation model is constructed based on the anatomical structure of the human airway, including a complete airway tree structure from the trachea to the bronchioles. The three-dimensional airway structure data is obtained through CT image reconstruction, accurately restoring the diameter, length, and bifurcation angle parameters of each level of bronchus.

[0019] Specifically, the fiber optic sensor array employs distributed fiber optic sensing technology, deploying a sensing point every 5 millimeters along the axial direction on the inner wall of the airway model. When the lens passes through, the sensor detects the pressure change and records the trigger moment. The instantaneous velocity of that segment is obtained by dividing the trigger time difference between two adjacent sensing points by the distance between the sensing points. The preset safe velocity range is set according to the anatomical characteristics of different airway segments, with the upper limit of the safe velocity for the main bronchus segment being higher than that for the segmental bronchi.

[0020] For example, the piezoelectric sensor is made of PVDF thin film material and is attached to a key location on the inner wall of the airway model. The tissue rebound time is measured as follows: after the lens applies pressure to cause radial compression deformation of the airway wall, the piezoelectric sensor continuously records the pressure change curve, starting from the moment the pressure is released and stopping when the pressure value recovers to more than 95% of the initial reference value. This time interval is the tissue rebound time. The standard rebound time threshold is obtained through experimental calibration, taking into account the influence of factors such as airway wall thickness, elastic modulus, and lumen diameter.

[0021] In one possible implementation, the strain gauge sensor built into the force feedback handle employs a full-bridge circuit configuration, capable of accurately measuring operating forces ranging from 0.1 Newtons to 50 Newtons. When the lens contacts the airway wall, generating an interaction force, this force is transmitted through the lens body to the handle end. The strain gauge deforms, causing a change in resistance, which is then converted into a feedback force intensity value by the signal conditioning circuit. The preset safety range lower limit is set based on the minimum sensing force threshold during normal operation.

[0022] It should be noted that by comprehensively judging data from three dimensions—exceeding speed limits, prolonged rebound time, and reduced feedback force—the system can accurately identify airway locations with operational risks, providing trainees with precise speed control guidance.

[0023] Step S102: Identify the tissue rebound time at the airway location where the lens advance speed exceeds the preset safe range, extract the high-risk tissue damage segment where the tissue rebound time deviation value exceeds the standard range, and visualize the high-risk tissue damage segment.

[0024] The system acquires airway location data where the lens advance speed exceeds a preset safe range, reads the corresponding tissue rebound time measurement, and calculates the difference between this value and a preset standard rebound time benchmark. When the difference exceeds a preset threshold of the benchmark, the airway location is marked as a preliminary risk point, resulting in a risk point dataset containing location coordinates and rebound time deviation values. Based on the rebound time deviation values ​​and location coordinates in the risk point dataset, a K-means clustering algorithm is used to group the risk points. During clustering, the deviation value is used as the primary feature, and the 3D coordinate distance is used as an auxiliary feature. When the coordinate distance between adjacent risk points is less than a preset distance threshold, they are considered spatially continuous. Risk points with similar deviation values ​​and spatially continuous locations are aggregated into high-risk tissue damage segments. The average deviation value and spatial range of each high-risk segment are calculated to obtain a boundary coordinate sequence of the high-risk segment. The boundary coordinate sequence is then used for region rendering on a 3D airway model. Different color intensities are assigned according to the average deviation value; high-risk segments with larger deviation values ​​are marked with a darker red. This is overlaid on the training interface to obtain a visualized distribution map of high-risk tissue damage segments.

[0025] For example, in one implementation, the identification of high-risk tissue damage areas is based on multi-dimensional data collected during bronchoscopy, and risk assessment is achieved through dynamic monitoring of tissue rebound time.

[0026] Specifically, the calculation of tissue rebound time deviation involves determining a standard baseline value, which is obtained through statistical analysis of bronchoscopic data from healthy volunteers. For different airway segments, such as the main bronchus, lobar bronchus, and segmental bronchus, corresponding standard rebound time databases are established. The baseline value for each airway segment considers factors such as diameter, wall thickness, and cartilage distribution density.

[0027] For example, the risk point dataset originates from the calculation of tissue rebound time deviation values ​​and the determination of baseline values, including the deviation values ​​and spatial coordinates of each risk point. When processing this dataset, the K-means clustering algorithm uses weighted Euclidean distance as a similarity metric. The weighting coefficients are dynamically adjusted according to the importance of the rebound time deviation values ​​and spatial coordinates, specifically through a function... Calculate the weight w of the deviation value d (where d is the current deviation value, max) d (where w is the maximum deviation value), the spatial distance weight is 1-w. d During the clustering process, the algorithm first randomly selects k initial cluster centers, where the value of k is determined by the elbow rule. By calculating the sum of squares within groups under different k values, the algorithm selects the k value corresponding to the inflection point of the curve. In each iteration, the algorithm calculates the weighted distance from each risk point to the cluster center, assigns the risk point to the category of the nearest cluster center, and then recalculates the center position of each category until the center position no longer changes significantly or the maximum number of iterations is reached.

[0028] Preferably, the spatial continuity is determined using the Manhattan distance in a three-dimensional coordinate system. When the sum of the coordinate differences between two risk points in the x, y, and z directions is less than 5 millimeters, they are considered spatially continuous. This method takes into account the tubular structure of the airway, avoiding misclassification of adjacent points on different branches as continuous segments.

[0029] In one possible implementation, the visualization rendering uses the HSV color space, where hue (H) is fixed at 0 degrees to represent red, saturation (S) is set to 100%, and lightness (V) is linearly mapped between 30% and 100% based on the deviation value. The larger the deviation value, the lower the lightness, and the deeper the red, thus visually reflecting the level of risk.

[0030] It should be noted that through the above-mentioned multi-dimensional data fusion and visualization, trainers can accurately identify high-risk areas in trainees' operations, adjust training strategies in a timely manner, and improve the relevance and safety of training.

[0031] Step S103: Extract the temporal fluctuation range of tissue compression and recovery in high-risk tissue damage areas, assess the difference between the temporal fluctuation range and the preset upper limit of deformation response time, and obtain a preliminary location record of tissue damage risk points.

[0032] The pressure application time and tissue morphology recovery time after pressure release are extracted from each sampling point in high-risk tissue damage areas. The duration of the compression phase and the duration of the recovery phase are calculated. Time data of multiple consecutive compression-recovery cycles are recorded, and the total duration of each cycle is calculated. The standard deviation of the total duration of multiple cycles is calculated to obtain the time fluctuation amplitude of tissue compression and recovery. The time fluctuation amplitude is compared with a preset upper limit of deformation response time, which is preset according to the tissue elasticity characteristics of different airway segments. If the fluctuation amplitude exceeds a preset proportional threshold of the upper limit, the sampling point is marked as a potential risk point, and the three-dimensional coordinates and the degree of fluctuation exceeding the limit of the potential risk point are recorded. The degree of fluctuation exceeding the limit of the potential risk point is used to classify the risk level. When the degree of exceeding the limit is in different preset ranges, it is marked as low, medium, and high risk levels, respectively. A spatial distribution map of risk points is constructed based on the risk level and three-dimensional coordinate information to obtain the preliminary location record of tissue damage risk points.

[0033] Specifically, in one implementation, the compression-recovery cycle refers to the cyclic process by which tissue recovers its original state after being subjected to compression stimulation, and its temporal stability reflects the repeatability of this process. Identification of tissue damage risk points is achieved by analyzing the temporal stability of this cycle, focusing on changes in the consistency of the tissue's response under repeated manipulations.

[0034] Specifically, the calculation of time fluctuation amplitude uses a sliding window method for real-time monitoring, based on pressure data collected by a pressure sensor during bronchoscope advancement. The sliding window comprises five consecutive compression and recovery cycles, with the total duration of each cycle defined as the time interval from the start of pressure application to complete tissue recovery. The pressure sensor records pressure changes at a sampling frequency of 100Hz. Compression begins when the pressure rises from zero to above a preset trigger threshold of 10Pa, and recovery is completed when the pressure recovers to within 5% of the initial baseline value and remains stable for more than 0.5 seconds. The standard deviation of the total duration of the five cycles within the window is calculated; this standard deviation represents the time fluctuation amplitude at the current sampling point. The fluctuation amplitude is updated each cycle the window advances.

[0035] Preferably, the upper limit of the deformation response time is set according to the anatomical location of the airway segment. For the main bronchus, due to its larger diameter and intact cartilage ring structure, the upper limit of the deformation response time is set to 2 seconds; for the lobar bronchus, the upper limit is set to 1.5 seconds; and for the segmental bronchus, due to its smaller diameter and softer tissue, the upper limit is set to 1 second. The preset proportional threshold is uniformly set to 80% of the upper limit value.

[0036] For example, the risk level classification adopts a three-level classification standard, which classifies risk levels based on the percentage threshold of the degree to which the volatility exceeds the limit, derived from an industry-standard risk assessment framework. Specifically, levels between 80% and 100% of the set upper limit are marked as low risk; levels between 100% and 120% are marked as medium risk; and levels exceeding 120% are marked as high risk. Each risk point record includes four attributes: three-dimensional coordinates, volatility value, degree of exceeding the limit, and risk level.

[0037] It should be noted that the spatial distribution map of risk points is presented using a point cloud visualization method. Points of different risk levels are marked with spheres of different sizes and colors: low risk is represented by small yellow spheres, medium risk by medium orange spheres, and high risk by large red spheres, enabling trainees to intuitively identify airway areas that require special attention.

[0038] Step S104: Analyze the preliminary location of the tissue damage risk point and record the corresponding feedback force intensity at the trainee's handle end. Compare it with the standard tactile feedback intensity benchmark of each airway segment in the training simulation model to identify risk points where the feedback force is lower than the standard benchmark. Assess the maximum withstand propulsion speed of the airway segment corresponding to the risk point, obtain the upper limit threshold of propulsion speed of the airway segment, and provide feedback to the trainee.

[0039] Preliminary location records of tissue damage risk points are obtained. The feedback force intensity value at the trainee's handle corresponding to each risk point location is read. The standard tactile feedback intensity benchmark for the corresponding airway segment is extracted from the training simulation model database. The difference between the measured feedback force and the standard benchmark is calculated. If the difference is negative and the absolute value exceeds a preset percentage threshold of the benchmark value, the point is marked as a feedback insufficiency risk point, resulting in a set of feedback insufficiency risk points. Based on the force deviation degree of each point in the feedback insufficiency risk point set, preset tissue attribute parameters of the corresponding airway segment are read from the training simulation model. A linear regression method is used to establish a mapping relationship with the force deviation value as the independent variable and the safe propulsion speed in the historical operation record as the dependent variable. The maximum withstand propulsion speed value of the airway segment where each feedback insufficiency risk point is located is calculated through the mapping relationship, obtaining the speed bearing capacity data of each risk point. Using the speed bearing capacity data, the minimum maximum withstand propulsion speed of multiple risk points within the same airway segment is taken as the preliminary safe propulsion speed of that airway segment. The preliminary safe propulsion speed is multiplied by a preset safety factor to obtain the upper limit threshold of the propulsion speed at the airway segment location. Speed ​​control prompts are generated by comparing the ratio of the upper limit threshold of propulsion speed to the current actual propulsion speed. When the ratio is less than a preset warning value, an audible alarm is triggered. At the same time, the relative relationship between the current speed and the upper limit threshold is displayed on the training interface with a color gradient bar, providing real-time feedback on speed control suggestions to the trainees.

[0040] Specifically, in one implementation, the feedback force assessment of the bronchoscopy training simulation model is based on multi-source data fusion. This fusion process includes collecting data from a force sensor and a propulsion speed sensor. First, the force deviation D and propulsion speed S are normalized to obtain dimensionless parameters D' and S', where D' = (DD / S') / S'. min ) / (D max -D min ), S'=(SS min ) / (S max -S min The normalized numerical range is [0,1]. The weighted average formula F=a×D'+b×S' is used, where F is the fusion feedback value, D' is the normalized force deviation, S' is the normalized propulsion speed, and a and b are weight coefficients, which are 0.6 and 0.4 respectively, satisfying a+b=1. This establishes a quantitative relationship between force deviation and propulsion speed, enabling real-time guidance for trainees' operations.

[0041] Specifically, the identification process for insufficient feedback risk points involves multi-level data comparison. Standard tactile feedback intensity benchmarks are stored and categorized according to the anatomical hierarchy of the airway: 8-12 Newtons for the main bronchus, 5-8 Newtons for lobar bronchioles, and 2-5 Newtons for segmental bronchioles. When the measured feedback intensity is lower than the lower limit of the corresponding airway segment's standard benchmark, and the deviation exceeds 20% of the benchmark value, that location is marked as a feedback insufficiency risk point. This grading standard takes into account the differences in tissue structure and operational difficulty among different airway segments.

[0042] For example, the construction of a linear regression model requires a large amount of historical operational data as training samples. The training simulation model automatically records key parameters during each operation, including lens advance speed, feedback force, tissue deformation, and operation duration. This data is stored as a time series, with each data point containing a timestamp, 3D coordinates, force value, and speed value. During the model training phase, operational data from historical records that did not result in tissue damage are extracted as positive samples; for example, at least 1000 sample points are selected. The force deviation values ​​from these positive samples are used as the independent variable x, and the corresponding advance speed is used as the dependent variable y. The least squares method is used to fit a linear relationship y = ax + b, where a is the slope coefficient and b is the intercept. and The calculation minimizes the sum of squared prediction errors. The linear regression model is established separately for different airway segments, with each segment having independent regression coefficients to accommodate differences in tissue characteristics at different sites. After the model is established, its predictive accuracy is evaluated through cross-validation. When the coefficient of determination R0 is... 2 A value greater than 0.8 indicates the model is usable.

[0043] Preferably, the historical operation records employ a circular buffer storage structure, retaining complete data from the most recent 1000 operations. Each record includes the operator's identification, operation time, operation trajectory, force and speed data at each sampling point, and an evaluation of the operation result. This data is periodically used to update the linear regression model, enabling the model to adapt to the operational characteristics of different trainees.

[0044] In one possible implementation, the calculation of the maximum withstand propulsion speed takes into account tissue fatigue effects, using a linear regression model to integrate fatigue coefficients. When the same airway segment experiences multiple compressions within a short period, its withstand capacity gradually decreases. The fatigue coefficient is determined based on the number of times the segment has been compressed in the past 5 minutes; the more compressions, the smaller the fatigue coefficient, thus correspondingly reducing the maximum withstand propulsion speed output by the regression model.

[0045] It should be noted that the safety factor is set using a tiered strategy, based on the assessment of trainees' experience levels and the risk calculation process.

[0046] Specifically, the safety factor k is determined by empirical grading: for beginners, k = 0.6, which is the upper limit of actual speed. in To maximize the propulsion speed, experienced trainees use k=0.7-0.8, adjusted based on training time and performance; advanced trainees, after multiple assessments, use k=0.9 to ensure safety matches ability. Furthermore, determining the upper speed limit threshold also needs to consider the continuity of adjacent airway segments. When multiple adjacent airway segments have risk points, the minimum upper speed limit threshold of these segments is taken as the uniform speed limit value for the entire area, avoiding frequent speed adjustments that could affect operational smoothness.

[0047] For example, speed control prompts use three feedback methods: visual, auditory, and tactile. Visual feedback is provided through a speed dashboard on the training interface, with the current speed indicated by a green pointer and the speed limit marked by a red scale; auditory feedback plays different frequencies of prompts depending on how close the speed is to the limit; tactile feedback reminds the trainee to adjust the operating force by changing the intensity of the handle vibration.

[0048] Understandably, by establishing a quantitative relationship between force deviation and propulsion speed, and combining it with a multi-layered safety protection mechanism, this solution can help trainees gradually improve their operational skills and form correct operating habits while ensuring training safety.

[0049] Step S105: Based on the upper limit threshold of the propulsion speed of the airway segment, evaluate the deviation from the current lens propulsion speed. When the upper limit threshold is lower than the current speed, generate the deceleration range to obtain the real-time speed limit threshold of each airway segment and display it in real time on the training interface.

[0050] The system obtains the upper limit threshold for propulsion speed at the airway segment location and the current lens propulsion speed. It calculates the difference between the current speed and the upper limit threshold. If the current speed exceeds the upper limit threshold, the speed reduction is determined based on the overspeed percentage. When the overspeed percentage is below a preset first threshold, the speed reduction is 10% of the current speed; between the first and second thresholds, the speed reduction is 20%; and when it exceeds the second threshold, the speed reduction is 30%, thus obtaining preliminary speed control parameters. Based on these preliminary speed control parameters and the difference in upper limit thresholds between adjacent airway segments, a moving average method is used to smooth the speed reduction. The moving window contains the speed reduction data from two sampling points before and after the current window. By calculating the arithmetic mean of the data within the window, abrupt changes in speed limits are avoided, resulting in a real-time speed limit threshold sequence for each airway segment. Using the real-time speed limit threshold sequence, a dynamic speed display layer is constructed on the training interface. Different speed limit levels are identified by color coding: green indicates normal speed, yellow indicates approaching the limit, and red indicates exceeding the limit. The relative positions of the current speed bar and the speed limit bar are displayed simultaneously. The display content is updated according to a preset refresh frequency to obtain the real-time speed limit threshold display results for each airway segment.

[0051] For example, in one implementation, the dynamic adjustment mechanism of the real-time speed limit threshold achieves continuity and stability of speed control by combining a multi-level deceleration strategy and a smoothing algorithm.

[0052] Specifically, the three-level classification of speed reduction is based on statistical analysis of a large amount of training data. When the overspeed percentage is less than 15%, it is considered slight overspeed, and the speed reduction is set at 10% of the current speed. This gentle adjustment allows trainees to maintain operational continuity. When the overspeed percentage is between 15% and 30%, it is considered moderate overspeed, and the speed reduction is increased to 20% to remind trainees to pay attention to control. When the overspeed percentage exceeds 30%, it is considered severe overspeed, and the speed reduction reaches 30%, forcing trainees to reduce their operating speed.

[0053] For example, the application of the moving average method takes into account the continuous movement of the bronchoscope within the airway. The moving window is designed to contain five consecutive sampling points: the current point and the deceleration amplitude data of the two sampling points before and after it. During the smoothing process, the arithmetic mean of all deceleration amplitude values ​​within the window is first calculated, and then this average is used as the actual deceleration amplitude at the current point. When the lens moves from the main bronchus into the lobar bronchus, the upper speed limits of the two airway segments may differ significantly. Through moving average processing, the change in speed limit is presented as a gradual process rather than an abrupt one, avoiding operational errors caused by sudden changes in speed limit.

[0054] Preferably, the real-time speed limit threshold sequence is updated at 10-millisecond intervals, and the limit threshold at each moment takes into account the characteristics of the current airway segment and historical operating trends, forming a dynamically adaptive speed control curve.

[0055] In one possible implementation, the interface displays an intuitive dashboard, with the outer ring of the speedometer dynamically adjusting its scale range based on the current airway segment's speed limit threshold. The green area represents the safe speed range, occupying 0-70% of the dial; the yellow area represents the warning speed range, occupying 70%-90%; and the red area represents the overspeed range, occupying 90%-100%. The current speed pointer is updated every 100 milliseconds based on real-time data.

[0056] It should be noted that by combining graded speed reduction and smoothing processing, a smooth transition in speed control is achieved, which improves the safety of training operations and the learning experience of trainees.

[0057] Step S106: Statistically analyze the actual propulsion speed of the trainee in each airway segment, compare the actual propulsion speed with the real-time speed limit threshold of each airway segment, determine the location of the airway segment that requires speed control, identify the distribution of airway segments where the trainee frequently violates the rules, analyze the trainee's operational weaknesses in speed control, tissue recognition, and force perception, and generate targeted suggestions for improving operational skills.

[0058] The actual propulsion speed sequences of trainees in each airway segment are extracted from the training record database. Real-time speed limit thresholds are read at corresponding time points, and the speed exceeding the limit value for each sampling point is calculated. When the exceeding value is greater than zero, it is marked as a violation point. The ratio of the number of violation points in each airway segment to the total number of sampling points is calculated to obtain the violation rate distribution data for each airway segment. Based on the violation rate distribution data, airway segments with violation rates exceeding a preset threshold are selected as high-frequency violation segments. Anatomical structural parameters corresponding to high-frequency violation segments are extracted from the airway model database as features. K-means clustering algorithm is used to classify high-frequency violation segments with similar features, obtaining the classification results for areas where trainees have difficulty operating. Using the violation feature data of each category in the classification results, corresponding speed control deviation values, tissue deformation response time deviations, and feedback force change rate data are extracted from the training records. The degree of difference between each dimension and the standard operating benchmark is calculated, and the dimension with the largest difference is identified as the main weakness of that category, obtaining the distribution of trainees' weaknesses in different operating dimensions. Based on the distribution of weaknesses, the system queries a preset training scheme data table according to the type and severity of weaknesses, selects the corresponding training module, and determines the training order by sorting the weaknesses from high to low. This generates targeted suggestions for improving operational skills, including training content, training order, and expected goals.

[0059] For example, in one implementation, the comprehensive evaluation of trainee operational behavior is based on full data recording during the training process. The training record database adopts a time-series database architecture, recording 100 sampling points per second, including multi-dimensional parameters such as timestamps, three-dimensional coordinates, propulsion speed, feedback force, and tissue deformation. The database is organized according to a three-level index of trainee identification, training session, and airway segment, supporting rapid batch data extraction and statistical analysis.

[0060] Specifically, the violation rate is calculated using a sliding time window method, with a window size of 10 seconds, sliding forward once per second. Within each window, the system compares the actual propulsion speed with the real-time speed limit threshold. When the actual speed exceeds the threshold, the excess value is calculated and the violation duration is accumulated. The violation rate is defined as the ratio of the violation duration to the total window duration. The violation rate for each airway segment is determined by averaging the violation rates of all windows within that segment. The threshold for identifying high-frequency violation segments is set at 30%, meaning segments with a violation rate exceeding 0.3 are marked as areas requiring close monitoring.

[0061] For example, the implementation of the K-means clustering algorithm includes three stages: feature extraction, distance calculation, and iterative optimization. In the feature extraction stage, anatomical parameters of each high-frequency violation segment are read from the airway model database to construct a feature vector. The feature vector contains five dimensions: airway diameter, wall thickness, bifurcation angle, radius of curvature, and distance from the carina (the distance from the airway initiation to the carina). These features are standardized to eliminate the influence of dimensions. Distance calculation uses Euclidean distance to measure the similarity between feature vectors. Initial cluster centers are selected using the K-means++ method (K-means++ is an optimization algorithm for selecting initial centers for K-means clustering), ensuring the dispersion of initial center points. During iterative optimization, each high-frequency violation segment is assigned to the category of the nearest cluster center, and then the center position of each category is recalculated as the mean of the feature vectors of all segments in that category. Iteration continues until the change in cluster centers is less than a preset threshold or the maximum number of iterations is reached. The number of clusters K is selected based on the silhouette coefficient, choosing the K value with the largest silhouette coefficient, typically between 3 and 5. The formula for calculating the silhouette coefficient is... , where s(i) is the silhouette coefficient of sample i, a(i) is the average distance from sample i to other samples in the same cluster, and b(i) is the average distance from sample i to the nearest cluster.

[0062] Preferably, anatomical parameters are automatically extracted from CT image data using three-dimensional reconstruction technology. The airway diameter is obtained by calculating the equivalent circular diameter of the cross-section, the wall thickness is determined by the average distance between the inner and outer contours, and the bifurcation angle is calculated by the angle between the centerlines of the parent bronchus and the daughter bronchus.

[0063] In one possible implementation, the three-dimensional evaluation vector is constructed by extracting features from the raw data in three dimensions: velocity, tissue response, and force. The velocity control deviation is obtained by calculating the root mean square error between the actual velocity and the ideal velocity trajectory. The ideal velocity trajectory is generated by fitting expert operational data using multinomial regression, a process that includes data cleaning, feature selection, and model validation. The tissue deformation response time deviation is determined by comparing the difference between the actual response time and the standard response time. The actual response time refers to the time required for tissue deformation to stabilize during operation, recorded in real time by sensors. The standard response time is calculated based on a mechanical model of the tissue's elastic modulus and the tube wall thickness, using the ratio of elastic modulus to thickness combined with an empirical coefficient. The feedback force change rate is obtained by performing a difference operation on the force time series, i.e., the difference between force values ​​at adjacent time points, and then calculating the standard deviation of the difference series, reflecting the stability of force control. The degree of difference in each dimension is expressed as a relative deviation percentage, i.e., the difference between the actual value (data measured during operation) and the benchmark value (expert or theoretical standard data) divided by the benchmark value. The dimensions with the largest relative deviations were identified as the main weaknesses, categorized as weak speed control, weak organizational perception, and weak force control, each corresponding to different training focuses.

[0064] It should be noted that the training plan data table is stored in a relational database and includes five fields: weakness type, severity, training module, training duration, and evaluation criteria. Each training module has a clear training objective and completion standard. For example, speed control training requires completing a specific trajectory within a specified time with a speed deviation not exceeding 10%.

[0065] For example, it is suggested that the training priority be determined by ranking the distribution of weaknesses based on their weights. The weakness score S and frequency F are normalized respectively, and the normalized weakness scores and frequencies are then weighted and summed to calculate the overall weight. In this embodiment, weaknesses with higher severity and frequency receive higher training priority. Based on the weaknesses identified by the 3D evaluation vector, 3-5 ordered combinations of training modules are generated after weight ranking, with the total training time and evaluation criteria for each module expected.

[0066] Understandably, through systematic data analysis and personalized suggestion generation, it is possible to accurately pinpoint trainees' operational weaknesses and provide targeted improvement solutions.

[0067] This invention provides a bronchoscopic operation training system for laparoscopic surgery, mainly comprising: The speed monitoring and tissue deformation sensing module is used to build a training simulation model. By installing the speed monitoring module and tissue deformation sensing module in the training simulation model, the lens advance speed, airway tissue rebound time and handle end feedback force intensity of the trainee's operation are obtained, and the airway position where the lens advance speed exceeds the preset safe range during the trainee's operation is identified. The high-risk tissue damage segment identification and visualization module is used to identify the tissue rebound time of airway locations where the lens advance speed exceeds the preset safety range, extract high-risk tissue damage segments where the tissue rebound time deviation value exceeds the standard range, and visualize the high-risk tissue damage segments. The preliminary location module for tissue damage risk points is used to extract the temporal fluctuation range of tissue compression and recovery in high-risk tissue damage areas, assess the difference between the temporal fluctuation range and the preset upper limit of deformation response time, and obtain a preliminary location record of tissue damage risk points. The feedback intensity analysis and propulsion speed upper limit assessment module is used to analyze the initial location record of tissue damage risk points and the corresponding feedback intensity of the student's handle. It is compared with the standard tactile feedback intensity benchmark of each airway segment in the training simulation model to identify risk points where the feedback intensity is lower than the standard benchmark. The module assesses the maximum propulsion speed that the airway segment corresponding to the risk point can withstand, obtains the upper limit threshold of the propulsion speed of the airway segment, and feeds it back to the student. The real-time speed limit threshold generation and display module is used to evaluate the deviation from the current lens advance speed based on the upper limit threshold of the advance speed of the airway segment. When the upper limit threshold is lower than the current speed, it generates the deceleration range, obtains the real-time speed limit threshold of each airway segment, and displays it in real time on the training interface. The module for analyzing operational weaknesses and generating skill improvement suggestions is used to statistically analyze the actual propulsion speed of trainees in each airway segment, compare the actual propulsion speed with the real-time speed limit threshold of each airway segment, determine the location of airway segments that require speed control, identify the distribution of airway segments where trainees frequently violate operating procedures, analyze the trainees' operational weaknesses in speed control, tissue recognition, and force perception, and generate targeted operational skill improvement suggestions.

[0068] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A bronchoscope operation training method for endoscopic surgery, characterized by, include: A training simulation model is constructed, in which a speed monitoring module and a tissue deformation sensing module are installed to acquire the lens advance speed, airway tissue rebound time, and handle feedback force intensity of the trainee's operation. The airway location where the lens advance speed exceeds the preset safety range is identified; the tissue rebound time of the airway location where the lens advance speed exceeds the preset safety range is identified, and high-risk tissue damage sections with tissue rebound time deviation values ​​exceeding the standard range are extracted and visualized on the training interface. Extract the time fluctuation amplitude of tissue compression and recovery in the high-risk tissue damage area, evaluate the difference between the time fluctuation amplitude and the preset upper limit of deformation response time, and obtain a preliminary location record of the tissue damage risk point; analyze the handle end feedback force intensity corresponding to the preliminary location record of the tissue damage risk point, compare it with the standard tactile feedback intensity benchmark of each airway segment in the training simulation model, identify risk points where the feedback force is lower than the standard benchmark, evaluate the maximum withstand propulsion speed of the airway segment corresponding to the risk point, obtain the upper limit threshold of propulsion speed of the airway segment position and feed it back to the trainee; based on the upper limit threshold of propulsion speed of the airway segment position, evaluate the deviation from the current lens propulsion speed, generate a deceleration amplitude when the upper limit threshold is lower than the current speed, obtain the real-time speed limit threshold of each airway segment and display it in real time on the training interface; count the trainee's actual propulsion speed in each airway segment, compare it with the real-time speed limit threshold of each airway segment, determine the airway segment position that requires speed control, identify the distribution of airway segments where the trainee frequently violates the rules, analyze the trainee's operational weaknesses in speed control, tissue recognition and force perception, and generate targeted operational skill improvement suggestions.

2. The bronchoscopic operation training method for laparoscopic surgery according to claim 1, characterized in that, The construction of the training simulation model includes installing a speed monitoring module and a tissue deformation sensing module to acquire the lens advance speed, airway segment tissue rebound time, and handle end feedback force intensity of the trainee's operation, and to identify airway locations where the lens advance speed exceeds the preset safe range. This includes: acquiring three-dimensional airway structure data of the bronchoscopy training simulation model; deploying fiber optic sensor arrays at the bifurcation locations of the main bronchus, lobar bronchus, and segmental bronchus to record the advance speed value and corresponding timestamp when the lens passes through each airway segment; calculating the instantaneous speed based on the trigger time difference of adjacent sensors; recording the airway segment position number when the speed exceeds the upper limit of the safe range; deploying a piezoelectric sensor at the airway segment position at the moment of speed exceeding the limit to collect the deformation of the airway wall after pressure; recording the time interval for the tissue to recover to its original shape after pressure release as the tissue rebound time; and using a strain gauge sensor built into the force feedback handle to measure the handle end feedback force intensity value.

3. The bronchoscopic operation training method for laparoscopic surgery according to claim 1, characterized in that, The process involves identifying the tissue rebound time at airway locations where the lens advance speed exceeds a preset safety range, extracting high-risk tissue damage segments where the tissue rebound time deviation exceeds a standard range, and visualizing these high-risk segments on a training interface. This includes: acquiring airway location data where the lens advance speed exceeds a preset safety range; reading the corresponding tissue rebound time measurement value; calculating the difference between this value and a preset standard rebound time benchmark value to obtain a risk point dataset containing location coordinates and rebound time deviation values; grouping these risk points using a K-means clustering algorithm, using deviation value as the primary feature and 3D coordinate distance as an auxiliary feature; aggregating risk points with similar deviation values ​​and continuous spatial locations into high-risk tissue damage segments; calculating the average deviation value and spatial range of each high-risk segment to obtain a boundary coordinate sequence for the high-risk segment; rendering these boundary coordinate sequences on a 3D airway model; assigning different color intensities based on the average deviation value; and overlaying these sequences on the training interface.

4. The bronchoscopic operation training method for laparoscopic surgery according to claim 1, characterized in that, Extract the time fluctuation amplitude of tissue compression and recovery in the high-risk tissue damage area, evaluate the difference between the time fluctuation amplitude and the preset upper limit of deformation response duration, and obtain a preliminary location record of the tissue damage risk point. This includes: extracting the pressure application time and the tissue morphology recovery time after pressure release from each sampling point in the high-risk tissue damage area, calculating the duration of the compression phase and the duration of the recovery phase, recording the time data of multiple consecutive compression and recovery cycles, calculating the total duration of each cycle and obtaining the standard deviation to obtain the time fluctuation amplitude value, comparing the time fluctuation amplitude value with the preset upper limit of deformation response duration, marking sampling points whose fluctuation amplitude exceeds the upper limit value by a preset proportion threshold as potential risk points, and recording the three-dimensional coordinates and fluctuation amplitude exceeding the limit value of the potential risk points.

5. The bronchoscopic operation training method for laparoscopic surgery according to claim 1, characterized in that, The process involves analyzing the initial location records of the tissue damage risk points and comparing them with the standard tactile feedback intensity benchmarks for each airway segment in the training simulation model. Risk points with feedback intensity lower than the standard benchmark are identified. The maximum allowable propulsion speed for the airway segment corresponding to the risk point is assessed, and the upper limit threshold for propulsion speed at the airway segment is obtained and fed back to the trainee. This includes: acquiring the initial location records of the tissue damage risk points; reading the feedback intensity value of the trainee's handle corresponding to each risk point location; calculating the difference between the measured feedback intensity and the standard benchmark; marking points with negative differences and absolute values ​​exceeding a preset percentage threshold of the benchmark as insufficient feedback risk points; reading preset tissue attribute parameters for the corresponding airway segment from the training simulation model; establishing a mapping relationship using a linear regression method with the force deviation as the independent variable and the safe propulsion speed in historical operation records as the dependent variable; calculating the maximum allowable propulsion speed value for the airway segment where each insufficient feedback risk point is located through the mapping relationship; taking the minimum maximum allowable propulsion speed of multiple risk points within the same airway segment and multiplying it by a preset safety factor to obtain the upper limit threshold for propulsion speed at the airway segment location.

6. The bronchoscopic operation training method for laparoscopic surgery according to claim 1, characterized in that, Based on the upper limit threshold of the propulsion speed at the location of the airway segment, the deviation from the current lens propulsion speed is evaluated. When the upper limit threshold is lower than the current speed, a reduction rate is generated. The real-time speed limit thresholds for each airway segment are obtained and displayed in real time on the training interface. This includes: obtaining the upper limit threshold of the propulsion speed at the location of the airway segment and the current lens propulsion speed value; calculating the difference between the current speed and the upper limit threshold; determining the reduction rate based on the overspeed percentage; combining the difference in the upper limit threshold of the speed between adjacent airway segments; smoothing the reduction rate using a moving average method; obtaining a real-time speed limit threshold sequence for each airway segment; constructing a dynamic speed display layer on the training interface; and using color coding to identify different speed limit levels.

7. The bronchoscopic operation training method for laparoscopic surgery according to claim 1, characterized in that, The actual propulsion speed of trainees in each airway segment is compared with the real-time speed limit threshold of each airway segment to determine the location of airway segments requiring speed control. The distribution of airway segments with high-frequency violations by trainees is identified, and the operational weaknesses of trainees in speed control, tissue recognition, and force perception are analyzed. Targeted suggestions for improving operational skills are generated, including: extracting the actual propulsion speed sequence of trainees in each airway segment from the training record database, calculating the speed exceedance value at each sampling point, statistically analyzing the violation rate distribution data of each airway segment, screening airway segments with violation rates exceeding a preset threshold as high-frequency violation segments, using the K-means clustering algorithm to classify high-frequency violation segments with similar characteristics, extracting the corresponding speed control deviation value, tissue deformation response time deviation, and feedback force change rate data, and identifying the dimension with the greatest difference as the main weakness.

8. A bronchoscopic operation training system for laparoscopic surgery, characterized in that, The system includes: A speed monitoring and tissue deformation sensing module is used to construct a training simulation model. The speed monitoring module and tissue deformation sensing module are installed in the training simulation model to obtain the lens advance speed, airway tissue rebound time and handle end feedback force intensity of the trainee's operation, and to identify the airway position where the lens advance speed exceeds the preset safe range. The high-risk tissue damage segment identification and visualization module is used to identify the tissue rebound time at the airway location where the lens advance speed exceeds the preset safety range, extract the high-risk tissue damage segment where the tissue rebound time deviation value exceeds the standard range, and visualize the high-risk tissue damage segment on the training interface. The preliminary location module for tissue damage risk points is used to extract the time fluctuation amplitude of tissue compression and recovery in the high-risk area of ​​tissue damage, evaluate the difference between the time fluctuation amplitude and the preset upper limit of deformation response time, and obtain a preliminary location record of the tissue damage risk point. The feedback intensity analysis and propulsion speed upper limit assessment module is used to analyze the initial location record of the tissue damage risk point and the corresponding handpiece end feedback intensity, compare it with the standard tactile feedback intensity benchmark of each airway segment in the training simulation model, identify risk points where the feedback intensity is lower than the standard benchmark, assess the maximum bearing propulsion speed of the airway segment corresponding to the risk point, obtain the propulsion speed upper limit threshold of the airway segment position and provide feedback to the trainee. The real-time speed limit threshold generation and display module is used to evaluate the deviation from the current lens advance speed based on the upper limit threshold of the advance speed of the airway segment position, generate the deceleration range when the upper limit threshold is lower than the current speed, obtain the real-time speed limit threshold of each airway segment, and display it in real time on the training interface. The module for analyzing operational weaknesses and generating skill improvement suggestions is used to statistically analyze the actual propulsion speed of trainees in each airway segment, compare it with the real-time speed limit threshold of each airway segment, determine the location of airway segments that require speed control, identify the distribution of airway segments where trainees frequently violate operating rules, analyze the trainees' operational weaknesses in speed control, tissue recognition, and force perception, and generate targeted operational skill improvement suggestions.

Images