Artificial intelligence-based surgical operation skill evaluation system and method

Abstract

The invention relates to the technical field of surgical operation skill evaluation systems, in particular to a surgical operation skill evaluation system and method based on artificial intelligence, and the system comprises a data collection module which collects image data in a surgical operation process; obtaining topological features of the image data; the feature processing module is electrically connected with the data acquisition module and receives the topological features sent by the data acquisition module; generating a feature matrix based on the topological features; performing group theory transformation on the feature matrix to obtain an enhanced feature matrix; the evaluation module is electrically connected with the feature processing module and receives the enhanced feature matrix sent by the feature processing module; based on the enhanced feature matrix, generating a skill score by using quantum probability and a chaotic power system; the output module is electrically connected with the evaluation module and is used for receiving the skill score sent by the evaluation module; and the skill score is output, and subtle differences and complex modes in the surgical operation can be captured, so that a more accurate and reliable evaluation result is provided.

Description

Technical Field

[0001] The present invention relates to the technical field of surgical operation skill evaluation systems, and in particular to a surgical operation skill evaluation system and method based on artificial intelligence. Background Art

[0002] With the continuous advancement of medical technology, the evaluation of surgical operation skills has received more and more attention. Traditional evaluation methods mainly rely on the subjective judgment of senior surgical experts. This method is not only time-consuming and labor-intensive, but also difficult to ensure the objectivity and consistency of the evaluation. In recent years, with the development of artificial intelligence technology, some surgical skill evaluation systems based on computer vision have begun to appear. These systems evaluate surgical skills by analyzing surgical videos and extracting some basic image features, such as instrument movement trajectory, operation speed, etc.

[0003] However, existing AI-based evaluation systems still have many shortcomings. First, these systems often only focus on the surface features of surgical operations, and it is difficult to capture the subtle details and complex spatial relationships during the operation. Secondly, the evaluation models of existing systems are relatively simple, and usually use traditional machine learning algorithms, which are difficult to cope with the high complexity and variability of surgical operations. In addition, these systems lack adaptive learning capabilities and cannot optimize the evaluation models based on continuously accumulated data. Finally, existing systems can usually only give simple scoring results, lack the explainability of the evaluation process, and cannot provide surgeons with targeted improvement suggestions. Summary of the invention

[0004] The present invention aims to solve the above-mentioned technical problems and provide a surgical operation skill evaluation system and method based on artificial intelligence, which can more accurately, objectively and efficiently evaluate surgical operation skills and provide surgeons with valuable feedback and improvement suggestions.

[0005] The present invention proposes a surgical operation skill evaluation system based on artificial intelligence, comprising:

[0006] Data acquisition module for:

[0007] Collect image data during surgical operations;

[0008] Acquiring topological features of the image data;

[0009] The feature processing module is electrically connected to the data acquisition module and is used to:

[0010] Receiving the topological features sent by the data acquisition module;

[0011] Based on the topological features, generating a feature matrix;

[0012] Performing group theory transformation on the characteristic matrix to obtain an enhanced characteristic matrix;

[0013] An evaluation module, electrically connected to the feature processing module, is used to:

[0014] Receiving the enhanced feature matrix sent by the feature processing module;

[0015] Based on the enhanced feature matrix, a skill score is generated using quantum probability and chaotic dynamical systems;

[0016] An output module, electrically connected to the evaluation module, is used to:

[0017] receiving the skill score sent by the evaluation module;

[0018] The skill score is output.

[0019] Preferably, the data acquisition module comprises:

[0020] An image acquisition unit, used for acquiring image data during surgical operation;

[0021] The topological feature extraction unit is electrically connected to the image acquisition unit and is used to extract topological features from the image data to obtain topological features.

[0022] Preferably, the feature processing module comprises:

[0023] A random matrix generating unit, used for generating a random matrix;

[0024] A characteristic matrix generating unit, electrically connected to the random matrix generating unit, and used for performing a Hadamard product operation on the topological feature and the random matrix to obtain a characteristic matrix;

[0025] The group theory transformation unit is electrically connected to the characteristic matrix generation unit and is used to perform group theory transformation on the characteristic matrix to obtain an enhanced characteristic matrix.

[0026] Preferably, the evaluation module comprises:

[0027] A quantum probability calculation unit, used for calculating a density operator based on the enhanced characteristic matrix;

[0028] A chaotic power system unit, electrically connected to the quantum probability calculation unit, and used to calculate the Lyapunov exponent based on the density operator;

[0029] A score generating unit is electrically connected to the chaotic power system unit and is used to generate a skill score based on the Lyapuno v index and probability distribution.

[0030] Preferably, the system further comprises a storage module, electrically connected to the evaluation module, for storing historical evaluation data and evaluation model parameters.

[0031] Preferably, the system further comprises a preprocessing module, which is arranged between the data acquisition module and the feature processing module, and is used for performing noise reduction, enhancement and standardization processing on the image data.

[0032] Preferably, the system further comprises a feedback module electrically connected to the evaluation module, for generating improvement suggestions according to the skill score.

[0033] Preferably, the system further comprises a human-computer interaction module electrically connected to the output module for displaying the skill scoring and evaluation process in a visual manner.

[0034] Preferably, the system further comprises an adaptive learning module electrically connected to the evaluation module for dynamically adjusting evaluation model parameters according to historical evaluation data.

[0035] The surgical operation skill assessment method based on artificial intelligence adopts the system and comprises the following steps:

[0036] Collect image data during surgical operations;

[0037] Acquiring topological features of the image data;

[0038] Based on the topological features, generating a feature matrix;

[0039] Performing group theory transformation on the characteristic matrix to obtain an enhanced characteristic matrix;

[0040] Based on the enhanced feature matrix, a skill score is generated using quantum probability and chaotic dynamical systems;

[0041] The skill score is output.

[0042] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0043] The present invention constructs a comprehensive and accurate surgical skill assessment framework by innovatively combining advanced technologies such as topological feature extraction, random matrix mapping, group theory transformation, quantum probability and chaotic dynamical systems. This multi-level and multi-dimensional assessment method can capture the subtle differences and complex patterns in surgical operations, thereby providing more accurate and reliable assessment results.

[0044] From a macro perspective, the system of the present invention realizes the automation of the entire process from data collection, feature processing to evaluation output, greatly improving the efficiency and objectivity of the evaluation. The modular design of the system enables each functional unit to work closely together to form an organic whole. For example, the collaborative work of the data acquisition module and the feature processing module ensures high-quality feature extraction; the seamless connection between the feature processing module and the evaluation module ensures the consistency and accuracy of the evaluation process.

[0045] At the microscopic level, there is also a deep synergy between the various technical innovations of the present invention. Topological feature extraction technology can effectively capture the spatial relationship in surgical operations, while random matrix mapping and group theory transformation further enhance the robustness and expressiveness of features. This feature processing method complements the subsequent quantum probability and chaotic dynamic system evaluation model, and together they construct an evaluation framework that can capture both details and the overall picture.

[0046] The present invention also cleverly resolves the contradiction between accuracy and efficiency. By adopting a highly parallelized algorithm design and an optimized data processing flow, the system provides high-precision evaluation while also achieving rapid response. This enables the system to provide evaluation results immediately after the operation, providing timely feedback to the doctor.

[0047] In addition, the adaptive learning module of the present invention forms a virtuous cycle with the evaluation module, continuously optimizing the evaluation model and improving the long-term performance of the system. This self-improvement mechanism ensures that the system can adapt to different types of surgeries and changing medical practices.

[0048] In general, the present invention has achieved a comprehensive improvement in evaluation accuracy, efficiency, interpretability and practicality through the organic combination of multiple technological innovations. This not only provides surgeons with an objective and reliable skill assessment tool, but also provides strong technical support for improving overall surgical quality, accelerating physician training and improving patient prognosis. With the widespread application of this system, it is expected to set off a revolutionary change in the field of surgery and promote the rapid development of precision medicine and personalized training. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a logic block diagram of the overall system of the present invention.

[0050] Figure 2 It is a logic block diagram of the data acquisition module of the present invention.

[0051] Figure 3 It is a logic block diagram of the feature processing module of the present invention.

[0052] Figure 4 It is a logic block diagram of the evaluation module of the present invention. DETAILED DESCRIPTION

[0053] Please refer to Figure 1-4 The present invention relates to a surgical operation skill evaluation system and method based on artificial intelligence. The system comprises a data acquisition module 1, a feature processing module 2, an evaluation module 3 and an output module 4.

[0054] The data acquisition module 1 is used to collect image data during surgical operations and obtain topological features of the image data. Preferably, the data acquisition module 1 includes an image acquisition unit 11 and a topological feature extraction unit 12. The image acquisition unit 11 can use a high-resolution camera to capture images during surgical operations in real time. The topological feature extraction unit 12 uses advanced computer vision algorithms to extract key topological features from the image. For example, a continuous homology algorithm can be used to extract topological invariants of the image, which can effectively characterize the spatial relationship between surgical instruments and tissues.

[0055] In the present invention, the data acquisition module 1 captures a continuous image sequence of the surgical operation in real time through a high-resolution camera. These image data contain rich spatial information and temporal dynamic information, which can reflect the complex relationship between surgical instruments, tissues and doctor operations. The persistent homology algorithm is used to extract topological invariants from the image. These invariants describe the spatial distribution and changes of surgical instruments and tissues, such as the changing trajectory of the contact point between the scalpel and the tissue, the shape of the suture line, etc. These topological features can characterize the key geometric structures and spatial relationships in the surgical operation, and provide important basic data for subsequent evaluation. For example, during the suturing process, the persistent homology algorithm can capture the motion trajectory of the suture needle and its interaction mode with the tissue, thereby quantifying the accuracy and stability of the surgical operation.

[0056] The feature processing module 2 is electrically connected to the data acquisition module 1, and is used to receive the topological features sent by the data acquisition module 1, generate a feature matrix based on the topological features, and perform group theory transformation on the feature matrix to obtain an enhanced feature matrix. In one embodiment of the present invention, the feature processing module 2 includes a random matrix generation unit 21, a feature matrix generation unit 22, and a group theory transformation unit 23.

[0057] The random matrix generation unit 21 generates a random matrix, which can increase the robustness of the feature. The feature matrix generation unit 22 performs a Hadamard product operation on the topological feature and the random matrix to obtain a feature matrix. This process can be expressed as:

[0058]

[0059] Where F is the feature matrix, is the topological feature of image I, R is a random matrix, and ⊙ represents the Hadamard product.

[0060] In surgery, This may include the boundaries of surgical instruments, morphological features of tissues, etc. The introduction of the random matrix R increases the system's noise resistance, enabling the system to better handle image noise caused by lighting changes or camera shake. The addition of the random matrix makes the system more stable to environmental changes (such as light changes, camera angle offset). The Hadamard product operation retains important information about the original topological features while enhancing the expressiveness of the features, which is helpful for subsequent group theory transformations. Through this feature enhancement method, the system can more accurately identify the operating details of surgical instruments, such as distinguishing between cutting actions and suturing actions, thereby providing a reliable basis for skill scoring.

[0061] The group theory transformation unit 23 performs group theory transformation on the feature matrix to obtain an enhanced feature matrix. This step can improve the invariance and expression ability of the feature. The group theory transformation can be expressed as:

[0062]

[0063] Among them, F enhanced is the enhanced feature matrix, G is the transformation group, g is the element in the group, Represented in Hilbert space The inner product operation in .

[0064] During surgical operations, images from different perspectives may cause changes in the feature matrix. For example, when surgical instruments approach tissue from different angles, their topological features may be different. Through group theory transformation, the system can eliminate the impact of perspective changes and extract more invariant features.

[0065] The group theory transformation makes the feature matrix more robust to changes in perspective and scale. By taking a weighted average of multiple transformed features, the system can more comprehensively describe the complexity of surgical operations. This transformation method is particularly suitable for surgical scenarios that require multi-perspective analysis, such as multi-angle observation of endoscopes in minimally invasive surgery. By enhancing the invariance of features, the system can more accurately evaluate the doctor's operating skills.

[0066] The evaluation module 3 is electrically connected to the feature processing module 2, and is used to receive the enhanced feature matrix sent by the feature processing module 2, and generate a skill score based on the enhanced feature matrix using quantum probability and chaotic dynamics. The present invention adopts this complex evaluation method to more accurately capture the subtle differences in surgical operations, thereby providing a more accurate skill evaluation.

[0067] The output module 4 is electrically connected to the evaluation module 3, and is used to receive and output the skill score sent by the evaluation module 3. The output module 4 can present the score to the user in an intuitive manner, such as in the form of a chart or a numerical value.

[0068] The system of the present invention achieves an accurate evaluation of surgical operation skills by combining advanced mathematical theories and artificial intelligence technology. The system can not only objectively evaluate surgical skills, but also provide surgeons with targeted improvement suggestions, thereby improving surgical quality and patient safety. The evaluation module 3 of the present invention includes a quantum probability calculation unit 31, a chaotic dynamic system unit 32 and a score generation unit 33. The quantum probability calculation unit 31 calculates the density operator based on the enhanced feature matrix. This process can be expressed as:

[0069]

[0070] Where, ρ represents the density operator; t represents the time parameter; represents the Liouville operator; F enhanced represents the enhanced feature matrix.

[0071] During surgical operations, density operators can be used to model the interaction between surgical instruments and tissues. For example, when a scalpel cuts tissue, the changes in its position, speed, and direction can be described by the dynamic equations of the density operator. Through the evolution equations of the density operator, the system can capture the dynamic characteristics of surgical operations, such as changes in instrument speed, tissue deformation, etc. The quantum probability model can more finely characterize the uncertainty of surgical operations, thereby improving the accuracy of the evaluation. The introduction of the density operator enables the system to model surgical operations in a quantum mechanical manner, which is particularly important for scenarios that require high-precision evaluation (such as neurosurgery).

[0072] The chaotic dynamic system unit 32 calculates the Lyapunov exponent based on the density operator. The calculation of the Lyapunov exponent can be expressed as:

[0073]

[0074] Among them, λ represents the Lyapunov exponent; P represents the probability distribution vector; δP(t) represents the disturbance at time t.

[0075] In surgical operations, the Lyapunov exponent can be used to evaluate the stability of the doctor's operation. For example, when the doctor's hand shakes during suturing, the system can quantify the impact of this shaking by calculating the Lyapunov exponent.

[0076] The Lyapunov index can directly reflect the stability of surgical operations and help identify potential risk factors. Through real-time monitoring of the Lyapunov index, the system can issue a warning when the operation is unstable, reminding the doctor to adjust the operation method. The application of the Lyapunov index enables the system to evaluate surgical operations from a dynamic perspective, which is particularly important for complex operations (such as heart surgery).

[0077] The score generating unit 33 generates a skill score based on the Lyapunov exponent and the probability distribution. The generation of the score can be expressed as:

[0078] S = σ(λ·P),

[0079] Among them, S represents the skill score; σ represents the Sigmoid function.

[0080] In skill score generation, the Sigmoid function combines the Lyapunov exponent and probability distribution to generate a comprehensive score. For example, when λ is smaller, the score will be higher, indicating that the operation is more stable.

[0081] By combining multiple indicators (such as stability and state distribution), the system is able to generate a more comprehensive skill score. The Sigmoid function converts complex mathematical calculation results into easy-to-understand numerical scores. The generation of skill scores not only provides doctors with objective evaluation results, but also provides improvement suggestions through the feedback module, thereby helping doctors continuously improve their operation level.

[0082] The present invention uses Gaussian filtering to remove noise from the image to ensure the accuracy of subsequent feature extraction. The image contrast is improved through histogram equalization to highlight the key details of surgical instruments and tissues. The image is scaled and normalized to make it suitable for a unified feature extraction algorithm. The spatial relationship between surgical instruments and tissues is extracted using a continuous homology algorithm to provide a geometric basis for skill assessment. The robustness of features is enhanced through random matrix technology to ensure the system's adaptability to environmental changes. Group theory transformation eliminates the impact of perspective changes and extracts more invariant features.

[0083] Through density operator calculation, the dynamic characteristics of surgical operations are modeled to capture the interaction between instruments and tissues. The stability of the operation is quantified through Lyapunov exponent calculation to identify potential risks. Finally, multiple indicators are combined to generate the final score to provide doctors with an objective evaluation.

[0084] The present invention achieves accurate evaluation of surgical operation skills by combining advanced mathematical theories (such as group theory, quantum probability, chaotic dynamic systems, etc.) and artificial intelligence technology. These effects are not only reflected in the accuracy of skill scoring, but also in the robustness, stability and applicability of the system. Through this comprehensive evaluation method, the system can help surgeons continuously improve their operation skills, thereby improving surgical quality and patient safety.

[0085] The system of the present invention further includes a storage module 5, which is electrically connected to the evaluation module 3 and is used to store historical evaluation data and evaluation model parameters. The storage module 5 can use a high-speed solid-state hard disk to ensure fast reading and writing of data. Preferably, the storage module 5 can also realize encrypted storage of data to protect sensitive medical information.

[0086] In one embodiment of the present invention, the system further includes a preprocessing module 6, which is arranged between the data acquisition module 1 and the feature processing module 2. The preprocessing module 6 is used to perform noise reduction, enhancement and standardization processing on the image data. For example, Gaussian filtering can be used for noise reduction, histogram equalization can be used for image enhancement, and image standardization can be achieved through scaling and normalization. These preprocessing steps can significantly improve the effect of subsequent feature extraction and processing.

[0087] The system of the present invention further includes a feedback module 7, which is electrically connected to the evaluation module 3 and is used to generate improvement suggestions based on the skill score. The feedback module 7 can provide targeted improvement suggestions to the surgeon based on the preset scoring criteria and expert knowledge base. For example, when the score of a certain operation skill is lower than a preset threshold (such as 80 points), the system can provide corresponding training suggestions or operation skills.

[0088] The present invention achieves accurate evaluation of surgical skills by combining quantum probability and chaotic dynamic systems. This innovative evaluation method can capture subtle differences that are difficult to identify with traditional methods, thereby providing more accurate and objective evaluation results. At the same time, the system of the present invention also has functions such as data storage, image preprocessing and feedback generation, forming a complete closed loop of skill evaluation and improvement. This can not only help surgeons objectively evaluate their own skills, but also provide them with targeted improvement suggestions, thereby continuously improving surgical quality and patient safety.

[0089] The system of the present invention also includes a human-computer interaction module 8, which is electrically connected to the output module 4 and is used to display the skill scoring and evaluation process in a visual manner. The human-computer interaction module 8 preferably uses a high-resolution touch screen to provide an intuitive and user-friendly interface. In a preferred embodiment of the present invention, the human-computer interaction module 8 can generate a multi-dimensional scoring chart, such as a radar chart or a bar chart, to intuitively display the skill level of various aspects of the surgical operation. In addition, the module can also replay key surgical operation clips and display the score changes in real time during the playback process to help surgeons better understand the evaluation results.

[0090] The system of the present invention also includes an adaptive learning module 9, which is electrically connected to the evaluation module 3 and is used to dynamically adjust the evaluation model parameters according to the historical evaluation data. The adaptive learning module 9 uses advanced machine learning algorithms, such as reinforcement learning or meta-learning, to continuously optimize the evaluation model. Preferably, the module can regularly analyze a large amount of historical evaluation data, identify potential deviations in the evaluation model, and automatically adjust the model parameters. For example, when it is found that the score of a certain type of surgical operation is significantly different from the expert evaluation, the system can automatically adjust the relevant weight coefficients. This adaptive learning mechanism ensures that the evaluation system can continuously improve its accuracy and applicability over time.

[0091] The present invention also provides a surgical operation skill assessment method based on artificial intelligence, comprising the following steps:

[0092] First, image data during the surgical operation is collected. In this step, a high-speed camera can be used to capture a continuous image sequence of the surgical operation at a speed of not less than 60 frames per second.

[0093] Secondly, the topological features of the image data are obtained. This step uses advanced computer vision algorithms, such as the continuous homology algorithm, to extract key topological features from the image. These features can effectively characterize the spatial relationship between surgical instruments and tissues and their changes.

[0094] Then, based on the topological features, a feature matrix is ​​generated. This step uses random matrix technology to enhance the robustness of the features through Hadamard product operations.

[0095] Next, the feature matrix is ​​subjected to group theory transformation to obtain an enhanced feature matrix. Group theory transformation can improve the invariance and expression ability of features.

[0096] Subsequently, based on the enhanced feature matrix, a skill score is generated using quantum probability and chaotic dynamics system, and the skill score is output. This step can present the score results in a variety of ways, such as numerical values, charts or detailed reports.

[0097] The method of the present invention combines advanced mathematical theory and artificial intelligence technology to achieve accurate and objective evaluation of surgical operation skills. This method can not only provide accurate skill scores, but also provide surgeons with targeted improvement suggestions, thereby continuously improving surgical quality and patient safety.

[0098] In order to verify the superiority of the surgical operation skill evaluation system and method based on artificial intelligence of the present invention, a set of simulation experiments was designed. Thirty surgeons with different skill levels were selected to perform laparoscopic cholecystectomy operations. The experimental conditions are as follows:

[0099] Surgical environment: Standardized laparoscopic surgery simulator;

[0100] Surgical object: Highly realistic human model;

[0101] Surgical instruments: standard laparoscopic surgical instrument set;

[0102] Duration of surgery: Each surgery is limited to 60 minutes;

[0103] We compared the embodiments of the present invention with two comparative examples:

[0104] Embodiment: Surgical operation skill evaluation system and method based on artificial intelligence using the present invention

[0105] Comparative Example 1: Traditional expert scoring method (3 senior surgical experts scored based on surgical videos)

[0106] Comparative Example 2: Evaluation system based on simple image processing (using only basic image features and traditional machine learning algorithms)

[0107] Evaluation indicators include: scoring accuracy, evaluation time, scoring consistency, and relevance of improvement suggestions. Scoring accuracy is obtained by comparing with the comprehensive scores of a team of senior experts; evaluation time is the time required for the system to give a final score; scoring consistency is calculated using the intraclass correlation coefficient (ICC); and the relevance of improvement suggestions is obtained by conducting a questionnaire survey on the doctors participating in the experiment.

[0108] The experimental results are shown in Table 1:

[0109] Table 1. Comparison of experimental results of Example 1, Comparative Example 1 and Comparative Example 2

[0110] Evaluation Methodology Rating accuracy Evaluation time Scoring consistency (ICC) Improve suggestion relevance Example 1 95.8％ 2 minutes 0.92 89％ Comparative Example 1 87.5％ 120 minutes 0.78 75％ Comparative Example 2 82.3％ 5 minutes 0.85 62％

[0111] It can be seen from the experimental results in Table 1 that the present invention has obvious advantages in all indicators. In particular, in terms of scoring accuracy, the present invention has reached 95.8%, which is much higher than the 87.5% of the traditional expert scoring method and the 82.3% of the simple image processing system. This result fully proves the remarkable effect of the innovative technologies such as topological feature extraction, random matrix mapping and group theory transformation adopted by the present invention in improving the accuracy of evaluation.

[0112] In terms of evaluation time, the present invention can complete the evaluation in just 2 minutes, which is much better than the 120 minutes required by the traditional expert scoring method and more efficient than the 5 minutes required by the simple image processing system. This high efficiency enables the present system to provide feedback immediately after the operation, which helps doctors improve their skills in a timely manner.

[0113] In terms of scoring consistency, the ICC of the present invention reaches 0.92, indicating that its evaluation results have high reliability and stability. This is due to the quantum probability and chaotic dynamic system evaluation method adopted by the system, which can capture the subtle differences in surgical operations, thereby providing more objective and consistent evaluation results.

[0114] The present invention also performed well in terms of the relevance of improvement suggestions, with a relevance of 89%, indicating that the system can provide surgeons with practical and useful improvement suggestions. This is mainly due to the system's adaptive learning module, which can continuously optimize the evaluation model and suggestion generation algorithm based on historical data.

[0115] The best implementation is to further optimize the system parameters based on the above experiments. Specifically, we adjusted the random matrix generation algorithm, adopted a method based on Wishart distribution, and introduced more complex Lie group transformations in group theory transformations. These optimizations further improved the scoring accuracy to 97.2%, shortened the evaluation time to 1.5 minutes, and improved the scoring consistency (ICC) to 0.95.

[0116] This set of experimental results fully demonstrates the innovation and superiority of the present invention in the field of surgical operation skill assessment. The system can not only provide more accurate, consistent and efficient assessments, but also provide surgeons with targeted improvement suggestions. This is of great significance for improving the overall quality of surgical operations, shortening the learning curve of doctors, and ultimately improving the safety of patients' operations.

[0117] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An artificial intelligence-based surgical operation skill evaluation system, characterized in that: include: Data acquisition module for: Collect image data during surgical operations; Acquiring topological features of the image data; The feature processing module is electrically connected to the data acquisition module and is used to: Receiving the topological features sent by the data acquisition module; Based on the topological features, generating a feature matrix; Performing group theory transformation on the characteristic matrix to obtain an enhanced characteristic matrix; An evaluation module, electrically connected to the feature processing module, is used to: Receiving the enhanced feature matrix sent by the feature processing module; Based on the enhanced feature matrix, a skill score is generated using quantum probability and chaotic dynamical systems; An output module, electrically connected to the evaluation module, is used to: receiving the skill score sent by the evaluation module; The skill score is output.

2. The system according to claim 1, characterized in that The data acquisition module comprises: An image acquisition unit, used for acquiring image data during surgical operation; The topological feature extraction unit is electrically connected to the image acquisition unit and is used to extract topological features from the image data to obtain topological features.

3. The system according to claim 1, characterized in that The feature processing module comprises: A random matrix generating unit, used for generating a random matrix; A characteristic matrix generating unit, electrically connected to the random matrix generating unit, and used for performing a Hadamard product operation on the topological feature and the random matrix to obtain a characteristic matrix; The group theory transformation unit is electrically connected to the characteristic matrix generation unit and is used to perform group theory transformation on the characteristic matrix to obtain an enhanced characteristic matrix.

4. The system according to claim 1, characterized in that The evaluation module includes: A quantum probability calculation unit, used for calculating a density operator based on the enhanced characteristic matrix; A chaotic power system unit, electrically connected to the quantum probability calculation unit, and used to calculate the Lyapunov exponent based on the density operator; A score generating unit is electrically connected to the chaotic power system unit and is used to generate a skill score based on the Lyapuno v index and probability distribution.

5. The system according to claim 1, characterized in that The system also includes a storage module, which is electrically connected to the evaluation module and is used to store historical evaluation data and evaluation model parameters.

6. The system according to claim 1, characterized in that The system also includes a preprocessing module, which is arranged between the data acquisition module and the feature processing module and is used for performing noise reduction, enhancement and standardization processing on the image data.

7. The system according to claim 1, characterized in that The system also includes a feedback module electrically connected to the evaluation module, and configured to generate improvement suggestions based on the skill score.

8. The system according to claim 1, characterized in that The system also includes a human-computer interaction module, which is electrically connected to the output module and is used to display the skill scoring and evaluation process in a visual manner.

9. The system according to claim 1, characterized in that The system also includes an adaptive learning module, which is electrically connected to the evaluation module and is used to dynamically adjust evaluation model parameters according to historical evaluation data.

10. A method for evaluating surgical operation skills based on artificial intelligence, using the system according to any one of claims 19, characterized in that: The following steps are involved: Collect image data during surgical operations; Acquiring topological features of the image data; Based on the topological features, generating a feature matrix; Performing group theory transformation on the characteristic matrix to obtain an enhanced characteristic matrix; Based on the enhanced feature matrix, a skill score is generated using quantum probability and chaotic dynamical systems; The skill score is output.

Images