Surgical operation teaching aided training system and method

Abstract

The invention relates to the technical field of teaching assistance, in particular to a surgical operation teaching auxiliary training system and method. The method comprises the steps that after multi-source operation action data is collected and preprocessed, a multi-dimensional data vector is constructed and serves as a disturbance source, and disturbance intensity is estimated; constructing an action section set based on the preprocessed multi-source surgical action data, and calculating the disturbance gravity center of an action section in combination with disturbance intensity; calculating the disturbance energy flux of the action section based on the disturbance intensity, and calculating local disturbance stability energy in combination with the disturbance gravity center of the action section; and based on the local disturbance stability energy, in combination with the preprocessed multi-source operation action data, calculating an action quality quantitative score and a comprehensive score of the action section, and generating an auxiliary suggestion. The problems that a traditional training evaluation system lacks fine-grained modeling for continuous operation actions, cannot uniformly analyze multi-modal operation behaviors, lacks quantifiable evaluation for operation stability and posture control and lacks pertinence for training auxiliary feedback are solved.

Description

Technical Field

[0001] This invention relates to the field of teaching aids technology, and in particular to a surgical teaching aid training system and method. Background Technology

[0002] Current surgical teaching primarily relies on traditional apprenticeship methods, video demonstrations, and hands-on practice with animals or simulation models. While these methods help establish basic knowledge and memorize surgical procedures to some extent, they face numerous technical bottlenecks in areas such as feedback mechanisms for key operational skills, visualization analysis of movement details, and adjustment of training paths to individual differences. Firstly, current mainstream training assessment systems are mostly based on offline scoring, coarse-grained observation of the entire process, or simply using video data for trajectory playback. This makes it difficult for these systems to uniformly analyze the trainee's multimodal behaviors during the procedure, including stability, posture control, strength coordination, and muscle activation, and to accurately identify abnormal fluctuations and instability in specific movement segments. Secondly, traditional training assessment systems generally neglect the dynamic, continuous, and multi-source coupling characteristics of surgical operations, abstracting complex surgical movements into simplified models, leading to teaching assessments that deviate from real-world scenarios. Especially when dealing with complex procedures such as suturing, puncture, and retraction, training assessment systems lack quantifiable process indicators and cannot provide training strategies with process tracking capabilities and iterative adjustments. Furthermore, current teaching feedback is mostly static and conclusive, lacking data-driven intelligent analysis and personalized intervention mechanisms, and unable to dynamically generate training plans based on students' specific performance.

[0003] In summary, traditional training and evaluation systems still suffer from problems such as a lack of fine-grained modeling of continuous surgical actions, an inability to uniformly analyze multimodal operational behaviors, a lack of quantifiable assessment of operational stability and posture control, and a lack of targeted and personalized training assistance feedback. Summary of the Invention

[0004] This invention provides a surgical teaching and training system and method to solve the technical problems of traditional training and evaluation systems, such as lack of fine-grained modeling of continuous surgical actions, inability to uniformly analyze multimodal operational behaviors, lack of quantifiable assessment of operational stability and posture control, and lack of targeted and personalized training and auxiliary feedback.

[0005] The present invention provides a surgical teaching and training system and method, which specifically includes the following technical solutions: A surgical teaching aid training method includes the following steps: S1. Collect multi-source surgical action data and preprocess it to obtain preprocessed multi-source surgical action data, and construct a multi-dimensional data vector; use the multi-dimensional data vector as a perturbation source and estimate the perturbation intensity; based on the preprocessed multi-source surgical action data, construct a set of action segments, and calculate the perturbation centroid of the action segments by combining the perturbation intensity. S2. Based on the perturbation intensity, calculate the perturbation energy flux of the action segment, and combine it with the perturbation centroid of the action segment to calculate the local perturbation stability energy; based on the local perturbation stability energy, combined with the preprocessed multi-source surgical action data, calculate the action quality quantification score of the action segment, and obtain a comprehensive score; based on the action quality quantification score and comprehensive score of the action segment, generate auxiliary suggestions.

[0006] Preferably, S1 specifically includes: Based on the preprocessed multi-source surgical action data, the Euclidean velocity of position change and the intensity of posture perturbation are calculated, and a force-muscle-motion joint activation term is constructed to estimate the perturbation intensity of the perturbation source, thus obtaining the perturbation intensity.

[0007] Preferably, S1 specifically includes: Based on the disturbance intensity and the absolute position vector of the disturbance source, the disturbance centroid of the action segment is calculated.

[0008] Preferably, S2 specifically includes: Based on the disturbance intensity, combined with the absolute position vector of the disturbance source, and by introducing the divergence operator, the disturbance energy flux of the action segment is calculated.

[0009] Preferably, S2 specifically includes: Based on the perturbation centroid of the action segment, the target deviation vector of the action segment is calculated. Combining the perturbation energy flux and stability decay term of the action segment, the local perturbation stability energy is calculated.

[0010] Preferably, S2 specifically includes: Based on the local perturbation stability energy and preprocessed multi-source surgical action data, positive vectorization index and negative penalty index are constructed to calculate the action quality quantitative score of the action segment; the action quality quantitative scores of the action segment are weighted and fused to obtain a comprehensive score.

[0011] Preferably, S2 specifically includes: The action quality quantification score of the action segment is compared with a preset threshold to obtain the action segment score; the comprehensive score is compared with a preset global threshold to obtain the comprehensive score; auxiliary suggestions are generated based on the action segment score and the comprehensive score.

[0012] A surgical teaching and training aid system includes the following components: The system includes a multi-source surgical motion acquisition module, a data preprocessing and perturbation source tensor construction module, a motion segmentation module, a perturbation intensity and perturbation centroid calculation module, an energy flux and local perturbation stability energy calculation module, a motion quality scoring module, and an auxiliary suggestion generation module. The multi-source surgical action acquisition module collects the surgical operation process of the trainee in real time to obtain multi-source surgical action data; the multi-source surgical action acquisition module is connected to the data preprocessing and perturbation source tensor construction module. The data preprocessing and perturbation source tensor construction module preprocesses the multi-source surgical action data to obtain preprocessed multi-source surgical action data; based on the preprocessed multi-source surgical action data, a multi-dimensional data vector is constructed, and the multi-dimensional data vector is used as a perturbation source; the data preprocessing and perturbation source tensor construction module is connected to the action segmentation module, the perturbation intensity and perturbation centroid calculation module, and the action quality scoring module, respectively. The action segmentation module segments the preprocessed multi-source surgical action data to obtain a set of action segments; the action segmentation module is connected to the disturbance intensity and disturbance centroid calculation module. The disturbance intensity and disturbance centroid calculation module estimates the disturbance intensity of the disturbance source after the disturbance source is established, and obtains the disturbance intensity. Based on the disturbance intensity and the set of action segments, the disturbance centroid of the action segment is calculated. The disturbance intensity and disturbance centroid calculation module is connected to the energy flux and local disturbance stability energy calculation module. The energy flux and local perturbation stability energy calculation module calculates the perturbation energy flux of the action segment based on the perturbation intensity and combined with the preprocessed multi-source surgical action data, by introducing a divergence operator; and obtains the local perturbation stability energy based on the perturbation centroid of the action segment, combined with the perturbation energy flux of the action segment and the attitude perturbation; the energy flux and local perturbation stability energy calculation module is connected to the action quality scoring module. The motion quality scoring module couples local perturbation stability energy with preprocessed multi-source surgical motion data to obtain a quantitative motion quality score for a motion segment; the quantitative motion quality scores of the motion segments are then weighted and fused to obtain a comprehensive score; the motion quality scoring module is connected to the auxiliary suggestion generation module. The auxiliary suggestion generation module compares the action quality quantification score of the action segment with a preset threshold to obtain the action segment score result; compares the comprehensive score with a preset global threshold to obtain the comprehensive score result; and generates auxiliary suggestions based on the action segment score result and the comprehensive score result.

[0013] The beneficial effects of the technical solution of the present invention are: 1. By constructing a unified perturbation source vector structure from the preprocessed multi-source surgical action data, the subtle dynamic changes during the surgical procedure are effectively preserved. Compared with traditional methods that rely on single-channel (such as video or trajectory) analysis, this significantly improves the robustness and completeness of the operational behavior modeling, enabling the surgical teaching and training system to maintain accurate analysis capabilities even when facing complex, variable, and high-precision surgical action scenarios.

[0014] 2. Using perturbation intensity as the core driving force, and combining spatial position, posture perturbation, and force-muscle-motion joint activation behavior, a high-dimensional perturbation model with physical interpretability was designed. By further introducing the concepts of perturbation centroid and perturbation energy flux, a comprehensive model of multiple factors in the surgical procedure was developed. This model can not only accurately identify unstable areas in local operations but also assess their potential impact on the overall quality of the operation. It breaks through the bottlenecks of traditional time series evaluation and fixed trajectory matching modes, and has stronger process interpretability and index sensitivity.

[0015] 3. By using existing action segment recognition algorithms (such as LoViT), the surgical procedure is divided into multiple action segments with teaching semantics (such as grasping, cutting, suturing, etc.), and the perturbation stability energy and action quality quantification score are calculated for each action segment. Compared with the traditional method based on global scoring, the operation process can be structurally evaluated and feedback can be given segment by segment. This enables the surgical teaching and training system to have a closed-loop capability of "diagnosis + localization + cause analysis". It is convenient for teachers or the system to implement precise intervention and training plan customization. While improving the training relevance, it can effectively avoid the problem of training ineffectiveness caused by global scoring masking local problems. Attached Figure Description

[0016] Figure 1 This is a structural diagram of a surgical teaching and training auxiliary system according to the present invention; Figure 2 This is a flowchart of a surgical teaching aid training method according to the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] The following description, in conjunction with the accompanying drawings, details the specific scheme of the surgical teaching aid training system and method provided by the present invention.

[0020] See attached document Figure 1 The diagram illustrates a structural diagram of a surgical teaching aid training system according to an embodiment of the present invention. The system includes the following components: The system includes a multi-source surgical motion acquisition module, a data preprocessing and perturbation source tensor construction module, a motion segmentation module, a perturbation intensity and perturbation centroid calculation module, an energy flux and local perturbation stability energy calculation module, a motion quality scoring module, and an auxiliary suggestion generation module. The multi-source surgical motion acquisition module uses wearable IMU (Inertial Measurement Unit), a 3D depth camera array, force-sensitive resistor patches, and other devices to collect real-time data on the surgical procedures of the trainee, obtaining multi-source surgical motion data. This data includes the spatial coordinates of the surgical tools, their attitude angles (Eulerian angles), applied forces, contact feedback signals (electromyographic activation values), and instantaneous linear acceleration at each moment. The multi-source surgical motion acquisition module is connected to the data preprocessing and perturbation source tensor construction module. The data preprocessing and perturbation source tensor construction module preprocesses the multi-source surgical action data to obtain preprocessed multi-source surgical action data; based on the preprocessed multi-source surgical action data, a multi-dimensional data vector is constructed, and the multi-dimensional data vector is used as a perturbation source; the data preprocessing and perturbation source tensor construction module is connected to the action segmentation module, the perturbation intensity and perturbation centroid calculation module, and the action quality scoring module, respectively. The action segmentation module segments the preprocessed multi-source surgical action data to obtain a set of action segments; the action segmentation module is connected to the disturbance intensity and disturbance centroid calculation module. The disturbance intensity and disturbance centroid calculation module estimates the disturbance intensity of the disturbance source after the disturbance source is established, and obtains the disturbance intensity. Based on the disturbance intensity and the set of action segments, the disturbance centroid of the action segment is calculated. The disturbance intensity and disturbance centroid calculation module is connected to the energy flux and local disturbance stability energy calculation module. The energy flux and local perturbation stability energy calculation module calculates the perturbation energy flux of the action segment based on the perturbation intensity and combined with the preprocessed multi-source surgical action data, by introducing a divergence operator; and obtains the local perturbation stability energy based on the perturbation centroid of the action segment, combined with the perturbation energy flux of the action segment and the attitude perturbation; the energy flux and local perturbation stability energy calculation module is connected to the action quality scoring module. The motion quality scoring module couples the local perturbation stability energy with multi-dimensional factors obtained from preprocessed multi-source surgical motion data to obtain a quantitative motion quality score for the motion segment; it then performs weighted fusion of the quantitative motion quality scores for the motion segment to obtain a comprehensive score; the motion quality scoring module is connected to the auxiliary suggestion generation module. The auxiliary suggestion generation module compares the quantitative score of the action quality of the action segment with a preset threshold to obtain the score result of the action segment; compares the comprehensive score with a preset global threshold to obtain the comprehensive score result; and generates auxiliary suggestions based on the score results of the action segment and the comprehensive score result using the existing remote teacher interface.

[0021] See attached document Figure 2 The diagram illustrates a flowchart of a surgical teaching aid training method according to an embodiment of the present invention, which includes the following steps: S1. Collect multi-source surgical action data and preprocess it to obtain preprocessed multi-source surgical action data, and construct a multi-dimensional data vector; use the multi-dimensional data vector as a perturbation source and estimate the perturbation intensity; based on the preprocessed multi-source surgical action data, construct a set of action segments, and calculate the perturbation centroid of the action segments by combining the perturbation intensity. Wearable IMU (Inertial Measurement Unit), 3D depth camera array, force-sensitive resistor patch, and other devices are used to collect multi-source surgical action data of trainees in real time. The multi-source surgical action data includes the spatial position coordinates, posture angles (Euler angles), applied forces, contact feedback signals (electromyographic activation values), and instantaneous linear acceleration of surgical tools at each moment. The multi-source surgical action data is preprocessed by noise reduction, cleaning, synchronization, standardization, and normalization to obtain preprocessed multi-source surgical action data. The preprocessing process uses techniques well known to those skilled in the art and will not be described in detail here. Furthermore, based on the preprocessed multi-source surgical action data, a multi-dimensional data vector is constructed. , And the multidimensional data vector is used as a perturbation source, where, It is the first A disturbance source at time The preprocessed spatial coordinates represent the absolute position of the operating component in three-dimensional space; It is the first A disturbance source at time The preprocessed attitude angles, expressed in Euler angles, include nutation angle, precession angle and rotation angle, used to indicate the orientation of the instrument or hand; It is the first A disturbance source at time The preprocessed instantaneous linear acceleration represents the rate change trend of the current operation. , where is the Euclidean norm of the instantaneous linear acceleration. , and At time respectively No. One disturbance source in , and Acceleration components along the axial direction; It is the first A disturbance source at time The pre-processed electromyographic activation value indicates the muscle load status. It is the first A disturbance source at time The force exerted on the tissue after pretreatment , , and At time respectively No. One disturbance source in , and Force component along the axial direction; This indicates the number of disturbance sources (i.e., the number of data sources being collected simultaneously). The length of the sampling time series; After the disturbance sources are established, the disturbance intensity of each disturbance source is estimated based on disturbance modeling concepts and multimodal nonlinear mapping, thus obtaining the disturbance intensity. , This indicates the first stage of the surgical procedure. A disturbance source at time The combined impact of the resulting spatial and attitude fluctuations; the specific formula is: ; in, It is the Euclidean velocity norm of positional change, used to capture spatial instability at the macroscopic level; It is the Euclidean velocity of position change; It is based on the sensitivity of attitude angle changes, i.e. attitude disturbance intensity, which is used to compensate for the insensitivity to the risk of directional operation and to provide the ability to characterize micro-rotational disturbances. The role of trigonometric functions is to amplify the sensitivity to small attitude changes (especially angle-related operations). For example, the impact of angle deviation on the organization is much greater than that of position deviation. Introducing a square term can prevent sign cancellation, and taking the square root can balance the scale order of magnitude. It can contribute nonlinear sensitivity to the disturbance intensity and is easy to couple with other physical quantities. It is a force-muscle-movement joint activation term used to express the operator's comprehensive behavioral level of force exertion at the current moment. The product represents the comprehensive activation intensity of the three action-related elements, which is used to reflect the operator's operational tension state. Introducing logarithmic transformation can compress excessively high values, suppress abnormal force-muscle-movement peaks that cause scoring bias, and improve the smoothness of data distribution. Furthermore, the preprocessed multi-source surgical action data is treated as an action sequence and segmented using existing action segment recognition algorithms (such as LoViT, which employs an end-to-end Transformer model) to obtain a set of action segments with boundary significance. Each action segment represents a teaching behavior unit, such as "clamping," "hooking," "sewing," or "cutting"; among which Indicates the total number of segments; Indicates the first A segment of action, or a unit of instructional behavior. Duration range , , They represent the first The start and end times of a segment of action; After obtaining the disturbance intensity of each disturbance source, the disturbance centroid of each motion segment is further calculated. The purpose is to characterize the spatial redistribution trend of all perturbations at the current moment, thereby reflecting the overall control state of each surgical operation. The perturbation centroid is defined as follows: ; in, , indicating the first A disturbance source at time The absolute position vector in three-dimensional space. Indicates transpose; It is the first A disturbance source at time The disturbance intensity is a scalar value; Indicates the first A disturbance source at time The perturbation intensity contributes to the spatial position of the perturbation source with weighted value according to its perturbation intensity. The numerator represents the sum of the spatial positions of all perturbation sources weighted according to their perturbation intensity, and the denominator is the normalization factor, thus obtaining the perturbation centroid of each action segment. The perturbation centroid serves as an important variable for subsequent trajectory stability and target offset rate.

[0022] S2. Based on the perturbation intensity, calculate the perturbation energy flux of the action segment, and combine it with the perturbation centroid of the action segment to calculate the local perturbation stability energy; based on the local perturbation stability energy, combined with the preprocessed multi-source surgical action data, calculate the action quality quantification score of the action segment, and obtain a comprehensive score; based on the action quality quantification score and comprehensive score of the action segment, generate auxiliary suggestions.

[0023] To analyze how energy propagates in space during surgery, the concept of perturbation energy flux is introduced to describe how perturbation behavior converges or diverges in space; Each action segment at time Disturbance energy flux Expressed as: ; in, Indicates the first A disturbance source at time The velocity vector, through the absolute position vector The difference method is a well-known technique in the art and will not be elaborated here; ⊗ is the tensor product symbol, which represents the fusion of multidimensional information in space; A third-order tensor is generated to describe the directional distribution of the perturbation energy flux in space; This is a divergence operator used to measure the degree of divergence of a vector field. The technical purpose of the above process is to characterize the convergence or discreteness of spatial perturbation sources, indirectly reflecting the stability of the operation and the accuracy of direction control. To further quantify stability, local perturbation stability energy is introduced. The overall energy fluctuation during the operation is described by integrating the deviation of the center of gravity trajectory, flux field intensity, and attitude perturbation amplitude. The specific formula is as follows: ; in, It is the first Local disturbance stability energy of each action segment; It is a time integral variable; It is the first The target deviation vector of each action segment is determined by comparing the perturbation centroid of the action segment with the standard reference coordinates of the target structure (such as the stitching point, entry point, etc.) preset according to expert experience. The comparison yields the result, i.e. The target deviation vector plays a crucial role in subsequent stability assessment; Indicates the first Each action segment at time The perturbation energy flux; Indicates the first Each action segment at time The flux field intensity; At any moment The local standard deviation of the center of gravity trajectory is calculated within a sliding window based on the expert experience method. This is a well-known technique in the art and will not be elaborated here. It is the first A disturbance source at time The angle of rotation; It is a nonlinear attitude disturbance. It is a stability decay term, used to nonlinearly suppress high-angle fluctuations to avoid sudden posture dominating the score too high; It describes the degree of combination between operational precision and energy release; in the denominator For time-based The stability penalty term, constructed from the square of the local standard deviation of the center of gravity trajectory, is used to measure the stability of the disturbed center of gravity during operation. This is a term for suppressing attitude fluctuations, used to control the nonlinear weights that affect the overall score from rapid rotation; Finally, based on biomechanical perturbation modeling theory and non-stationary dynamic signal analysis methods, the local perturbation stability energy and multidimensional factors obtained from preprocessed multi-source surgical action data are coupled and processed to obtain the first... Quantitative scoring of the quality of each action segment The specific formula is as follows: ; in, For the first The operation direction offset angle of each action segment is used to reflect the correctness of the operation direction. It is derived from the IMU attitude data in the preprocessed multi-source surgical action data. This is a method well known to those skilled in the art and will not be described in detail here. Indicates the first The energy fluctuation rate of each action segment represents the volatility of the disturbance intensity. The larger the energy fluctuation rate, the more violent the fluctuation and the higher the penalty. It is calculated based on the disturbance intensity, which is a well-known technical method in the art and will not be described in detail here. This is the energy fluctuation sensitivity coefficient, used to control the degree of exponential penalty imposed by energy fluctuation rate on the quantitative scoring of movement quality. It is determined based on expert experience, with a reference value range of [value missing]. ; Indicates the first The maximum perturbation intensity of each action segment is used to capture the instability during intense operations. It is the first The disturbance oscillation of the first action segment indicates the first action segment. Whether there are drastic and rapid changes in each action segment, i.e. whether there are frequent "jittering" or "twitching" behaviors, is the sum of the integrals of the squares of the second-order difference of the disturbance intensity. The calculation method is a well-known technique in the art and will not be elaborated here. Indicates a positive vectorization index; This represents the inverse penalty index; finally, by using the cubic root, the quantitative score of action quality is made closer to the linear score, avoiding an overly steep score gradient. Furthermore, the first Quantitative scoring of the quality of each action segment Compared with the threshold preset by expert experience method By comparing, we obtain the first The scoring results of each action segment, that is, when If the quality of the movement is within the acceptable range as recognized by experts, it is considered to have met the standard; if the quality is within the acceptable range as recognized by experts, it is considered to have met the standard. If the deviation is too large, the stability is insufficient, or the posture is unreasonable, it is considered to be substandard. In addition, in the system implementation, Each component (such as) , , and These are explicitly recorded as sub-indicators. When a sub-indicator exceeds a preset threshold based on expert experience, the corresponding suggested semantic template is automatically triggered, thereby providing personalized auxiliary training suggestions based on data analysis for each specific action. For example, when... When the preset threshold is exceeded, suggestions are automatically generated, such as "Please pay attention to the control of the instrument angle to avoid deviation from the operation direction"; Furthermore, based on the pre-defined weights of each action segment according to expert experience, the quantitative scores of the action segment's quality are weighted and fused to obtain a comprehensive score. The comprehensive score will be compared with a global threshold preset based on expert experience. By comparing the results, a comprehensive score is obtained, that is, when When the overall operation quality is acceptable, the operation performance needs to be optimized. Finally, based on the scoring results of each action segment and the overall scoring results, auxiliary suggestions are provided using the existing remote teacher assistance interface.

[0024] In summary, a surgical teaching and training system and method have been developed.

[0025] The order of the embodiments is for illustrative purposes only and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0026] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0027] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for assisting in surgical teaching and training, characterized in that, Includes the following steps: S1. Collect multi-source surgical action data, perform preprocessing to obtain preprocessed multi-source surgical action data, and construct a multi-dimensional data vector; Multidimensional data vectors are used as disturbance sources, and the disturbance intensity is estimated. Based on the preprocessed multi-source surgical action data, a set of action segments is constructed, and the perturbation centroid of the action segments is calculated by combining the perturbation intensity. S2. Based on the disturbance intensity, calculate the disturbance energy flux of the action segment, and combine it with the disturbance centroid of the action segment to calculate the local disturbance stability energy; Based on the local perturbation stability energy and combined with preprocessed multi-source surgical action data, the action quality quantification score of the action segment is calculated, and a comprehensive score is obtained; based on the action quality quantification score and the comprehensive score of the action segment, auxiliary suggestions are generated.

2. The surgical teaching aid training method according to claim 1, characterized in that, S1 specifically includes: Based on the preprocessed multi-source surgical action data, the Euclidean velocity of position change and the intensity of posture perturbation are calculated, and a force-muscle-motion joint activation term is constructed to estimate the perturbation intensity of the perturbation source, thus obtaining the perturbation intensity.

3. The surgical teaching aid training method according to claim 2, characterized in that, S1 specifically includes: Based on the disturbance intensity and the absolute position vector of the disturbance source, the disturbance centroid of the action segment is calculated.

4. The surgical teaching aid training method according to claim 1, characterized in that, S2 specifically includes: Based on the disturbance intensity, combined with the absolute position vector of the disturbance source, and by introducing the divergence operator, the disturbance energy flux of the action segment is calculated.

5. The surgical teaching aid training method according to claim 4, characterized in that, S2 specifically includes: Based on the perturbation centroid of the action segment, the target deviation vector of the action segment is calculated. Combining the perturbation energy flux and stability decay term of the action segment, the local perturbation stability energy is calculated.

6. The surgical teaching aid training method according to claim 5, characterized in that, S2 specifically includes: Based on the local perturbation stability energy and preprocessed multi-source surgical action data, positive vectorization index and negative penalty index are constructed to calculate the action quality quantitative score of the action segment; the action quality quantitative scores of the action segment are weighted and fused to obtain a comprehensive score.

7. The surgical teaching aid training method according to claim 6, characterized in that, S2 specifically includes: The action quality quantification score of the action segment is compared with a preset threshold to obtain the action segment score; the comprehensive score is compared with a preset global threshold to obtain the comprehensive score; auxiliary suggestions are generated based on the action segment score and the comprehensive score.

8. A surgical teaching and training auxiliary system, applied to the surgical teaching and training auxiliary method as described in claim 1, characterized in that, Includes the following parts: The system includes a multi-source surgical motion acquisition module, a data preprocessing and perturbation source tensor construction module, a motion segmentation module, a perturbation intensity and perturbation centroid calculation module, an energy flux and local perturbation stability energy calculation module, a motion quality scoring module, and an auxiliary suggestion generation module. The multi-source surgical action acquisition module collects the surgical operation process of the trainee in real time to obtain multi-source surgical action data; the multi-source surgical action acquisition module is connected to the data preprocessing and perturbation source tensor construction module. The data preprocessing and perturbation source tensor construction module preprocesses the multi-source surgical action data to obtain preprocessed multi-source surgical action data; based on the preprocessed multi-source surgical action data, a multi-dimensional data vector is constructed, and the multi-dimensional data vector is used as a perturbation source; the data preprocessing and perturbation source tensor construction module is connected to the action segmentation module, the perturbation intensity and perturbation centroid calculation module, and the action quality scoring module, respectively. The action segmentation module segments the preprocessed multi-source surgical action data to obtain a set of action segments; the action segmentation module is connected to the disturbance intensity and disturbance centroid calculation module. The disturbance intensity and disturbance centroid calculation module estimates the disturbance intensity of the disturbance source after the disturbance source is established, and obtains the disturbance intensity. Based on the disturbance intensity and the set of action segments, the disturbance centroid of the action segment is calculated. The disturbance intensity and disturbance centroid calculation module is connected to the energy flux and local disturbance stability energy calculation module. The energy flux and local perturbation stability energy calculation module calculates the perturbation energy flux of the action segment based on the perturbation intensity and combined with the preprocessed multi-source surgical action data, by introducing a divergence operator; and obtains the local perturbation stability energy based on the perturbation centroid of the action segment, combined with the perturbation energy flux of the action segment and the attitude perturbation; the energy flux and local perturbation stability energy calculation module is connected to the action quality scoring module. The motion quality scoring module couples local perturbation stability energy with preprocessed multi-source surgical motion data to obtain a quantitative motion quality score for a motion segment; the quantitative motion quality scores of the motion segments are then weighted and fused to obtain a comprehensive score; the motion quality scoring module is connected to the auxiliary suggestion generation module. The auxiliary suggestion generation module compares the quantitative score of the action quality of the action segment with a preset threshold to obtain the score result of the action segment; The overall score is compared with a preset global threshold to obtain the overall score result; auxiliary suggestions are generated based on the score results of the action segments and the overall score result.

Images