Artificial intelligence-assisted diagnosis method and system for knee joint

Abstract

The present application provides an artificial intelligence-assisted diagnosis method and system for the knee joint, wherein multi-angle dynamic X-ray images and six-degree-of-freedom contact force distribution data in the joint cavity are collected synchronously, the patellar slip trajectory is dynamically tracked and a mechanical phase trajectory is generated; a personalized knee joint model is constructed in combination with the patient's lower limb biological parameters, and the mechanical phase trajectory is superimposed with the contact force data vector to generate a dynamic mechanical fingerprint, which is then mapped into a biomechanical property map; the time-frequency characteristics of the electromyographic signal and the tibial rotation angle during movement are collected to construct an electromyographic activation pattern, which is then fused with the biomechanical property map to form a comprehensive joint dynamics profile; the abnormal joint dynamics propagation model is used to analyze the stress wave conduction path to identify compensatory injury patterns; a diagnostic map is generated based on the recognition results, the diagnostic map is corrected by matching the preset biomechanical constraints, and an individualized rehabilitation strategy is formulated. The present application improves the treatment effect and the patient's recovery speed.

Description

Technical Field

[0001] The embodiments of the present application relate to the interdisciplinary field of biomedical engineering and artificial intelligence, and in particular to an artificial intelligence-assisted diagnosis method and system for knee joints. Background Art

[0002] With the aging of the global population and increased participation in sports activities, the incidence of knee-related diseases has climbed annually, creating an increasingly urgent need for early, accurate diagnosis and personalized treatment plans. Especially among athletes and the elderly, knee injuries caused by high-intensity exercise or the natural aging process not only severely impact patients' quality of life but can also lead to long-term health problems. To better address this challenge, the medical field requires a technology that can integrate multi-source data (such as dynamic X-ray images, intra-articular mechanical distribution, surface electromyography signals, etc.) and perform efficient comprehensive analysis. This technology should be able to deeply understand the complex biomechanical characteristics of the knee joint, provide accurate injury pattern recognition, and develop rehabilitation strategies that are tailored to individual differences to improve treatment outcomes and quality of life.

[0003] Existing technologies have attempted to utilize advanced biomechanical modeling and computer-aided analysis methods to assess the functional status of the knee joint. For example, by combining high-resolution medical imaging with dynamic X-ray imaging sequences, finite element analysis is used to construct a three-dimensional knee joint model and simulate its stress distribution under different load conditions. In addition, by collecting surface electromyographic signals and gait analysis to capture muscle activity and joint movement patterns, more comprehensive data support is provided for the biomechanical assessment of the knee joint. The application of these technologies has significantly improved the understanding and diagnosis of knee injuries, making treatment plans for specific cases more scientific and reasonable.

[0004] However, despite the progress made in these approaches, they still face several key challenges in practical applications: First, insufficient data fusion still exists, making it difficult to effectively integrate multiple data types such as imaging, mechanics, and electrophysiology, and thus unable to fully reflect the true condition of the knee joint; second, existing models fail to fully consider individual differences, and the degree of personalization is limited, which affects the accuracy of diagnosis; finally, the level of automation and intelligence still needs to be improved, and most systems still require manual intervention to interpret results and formulate rehabilitation strategies. Summary of the Invention

[0005] The embodiments of the present application provide an artificial intelligence-assisted diagnosis method and system for the knee joint, which are used to solve the problems of poor treatment effect and slow patient recovery in the prior art.

[0006] In a first aspect, an embodiment of the present application provides an artificial intelligence-assisted diagnosis method for a knee joint, comprising:

[0007] receiving a multi-angle dynamic X-ray image sequence and six-degree-of-freedom contact force distribution data within the joint cavity, dynamically tracking the multi-angle dynamic X-ray image sequence and combining it with the six-degree-of-freedom contact force distribution data within the joint cavity, encoding the patellar sliding trajectory to generate mechanical phase trajectory data;

[0008] Collecting biological parameters of the patient's lower limbs to construct a personalized knee joint model, performing vector superposition of the mechanical phase trajectory data and the six-degree-of-freedom contact force distribution data within the joint cavity to generate a dynamic mechanical fingerprint, and mapping the dynamic mechanical fingerprint into a biomechanical property map using an inverse dynamics solver;

[0009] The frequency characteristics of the patient's surface electromyographic signals and the tibial rotation angle during exercise are collected to construct an electromyographic activation pattern marked with the exercise phase, and the biomechanical property map is fused with the electromyographic activation pattern to generate a comprehensive profile of joint dynamics;

[0010] The joint dynamics comprehensive profile is input into a pre-built joint abnormal dynamics propagation model, and the stress wave conduction equation is used to calculate the wave propagation path of the meniscus stress concentration area and the spatiotemporal evolution trajectory of the cruciate ligament deformation energy field. The compensatory injury pattern is identified by analyzing the interference effect between the spatiotemporal evolution trajectory and the wave propagation path;

[0011] According to the compensatory injury pattern, combined with the biomechanical property map, the myoelectric activation pattern and the comprehensive profile of joint dynamics, a diagnostic map is generated, the biomechanical constraints in the preset motion feature library are matched, and the diagnostic map is adaptively modified to generate an individualized rehabilitation strategy.

[0012] Optionally, the step of inputting the joint dynamics comprehensive profile into a pre-built joint abnormal dynamics propagation model, calculating the wave propagation path of the meniscus stress concentration area and the spatiotemporal evolution trajectory of the cruciate ligament deformation energy field using the stress wave conduction equation, and identifying the compensatory injury pattern by analyzing the interference effect between the spatiotemporal evolution trajectory and the wave propagation path includes:

[0013] Performing hierarchical tensor decomposition on the comprehensive joint dynamics profile to obtain biomechanical tensor components and kinematic tensor components, constructing a meniscus stress propagation path based on a stress waveguide model for the biomechanical tensor components, and calculating dynamic correlation parameters between the fluctuation attenuation rate and the stress concentration factor in the meniscus stress propagation path;

[0014] Establishing a ligament energy field evolution model of the viscoelastic deformation field for the kinematic tensor component, and extracting the spatiotemporal gradient and energy residence node coordinates of the ligament energy field evolution model;

[0015] An interference field equation is established by dynamically coupling a stress waveguide model with a viscoelastic deformation field using the dynamic correlation parameters, the spatiotemporal gradient, and the energy retention node coordinates; a wavefield interference intensity distribution between the meniscus stress propagation path and the energy retention node coordinates is calculated based on the interference field equation; and a compensatory damage pattern vector is generated based on the three-dimensional coordinates of an abnormal resonance region in the wavefield interference intensity distribution and an energy overflow threshold;

[0016] The compensatory injury pattern vector is modally matched with standard motion chain parameters in a motion feature library, and the compensatory injury pattern vector is subjected to frequency domain weighted adjustment based on the matching result to generate a compensatory injury pattern.

[0017] Optionally, the method includes establishing an interference field equation by dynamically coupling a stress waveguide model with a viscoelastic deformation field using the dynamic correlation parameters, the spatiotemporal gradient, and the energy retention node coordinates, calculating a wavefield interference intensity distribution between the meniscus stress propagation path and the energy retention node coordinates according to the interference field equation, and generating a compensatory damage pattern vector according to the three-dimensional coordinates of an abnormal resonance region in the wavefield interference intensity distribution and an energy overflow threshold, including:

[0018] Performing wave equation parameterized reconstruction on the meniscus stress propagation path, extracting the wave phase delay parameter and amplitude attenuation rate in the meniscus stress propagation path, performing energy gradient field decomposition on the energy residence node coordinates, and generating an energy density gradient vector and a residence time coefficient;

[0019] A dynamic coupling kernel function is constructed based on the wave phase delay parameter and the energy density gradient vector, and a spatial grid map of the wavefield interference intensity distribution is generated by combining the dynamic coupling kernel function with a spatial convolution operation, and grid cells with energy density peaks higher than an overflow threshold are marked in the spatial grid map as abnormal resonance regions;

[0020] Using the amplitude attenuation rate and the residence time coefficient, the product of the energy residence time coefficient and the fluctuation amplitude attenuation rate in the abnormal resonance area is calculated as the compensatory damage intensity factor, and the angle between the three-dimensional coordinates of the abnormal resonance area and the tangent direction of the fluctuation propagation path is extracted as the damage space distribution parameter. The compensatory damage intensity factor and the damage space distribution parameter are tensor-recombined to generate a compensatory damage pattern vector.

[0021] Optionally, constructing a dynamic coupling kernel function based on the wave phase delay parameter and the energy density gradient vector, generating a spatial grid map of the wavefield interference intensity distribution by combining the dynamic coupling kernel function with a spatial convolution operation, and marking grid cells with energy density peaks higher than an overflow threshold in the spatial grid map as abnormal resonance regions, includes:

[0022] Performing time series normalization processing on the fluctuation phase delay parameter to generate a phase delay weight coefficient, and performing orthogonal basis decomposition on the energy density gradient vector to obtain a normal energy gradient component and a tangential energy diffusivity;

[0023] A dual-channel coupling kernel function is constructed based on the phase delay weight coefficient and the normal energy gradient component, the tangential energy diffusivity is used as a dynamic attenuation factor of the dual-channel coupling kernel function, a three-dimensional convolution kernel is generated according to the dynamic attenuation factor, a dynamic window parameter of the spatial convolution operation is defined according to the anatomical structure of the knee joint, and a sliding integral calculation is performed in the biomechanical spatial domain using the dynamic window parameter in combination with the three-dimensional convolution kernel to generate an interference intensity distribution matrix;

[0024] The interference intensity distribution matrix is ​​subjected to multi-scale Gaussian filtering, the spatial topological structure of the energy density peak in the filtered matrix is ​​detected, and the topological units in the spatial topological structure that are continuously adjacent and have peak values ​​higher than an overflow threshold are clustered as abnormal resonance regions.

[0025] Optionally, constructing a dual-channel coupling kernel function based on the phase delay weight coefficient and the normal energy gradient component, using the tangential energy diffusivity as a dynamic attenuation factor of the dual-channel coupling kernel function, generating a three-dimensional convolution kernel based on the dynamic attenuation factor, defining dynamic window parameters of the spatial convolution operation based on the anatomical structure of the knee joint, and performing a sliding integral calculation in the biomechanical spatial domain using the dynamic window parameters in combination with the three-dimensional convolution kernel to generate an interference intensity distribution matrix, including:

[0026] Performing a frequency domain transform on the phase delay weight coefficient to generate a phase delay spectrum, extracting the main frequency band energy ratio in the phase delay spectrum as the time modulation factor of the dual-channel coupling kernel function, performing a spatial Fourier transform on the normal energy gradient component to generate a normal energy wavenumber spectrum, and extracting the high-frequency attenuation slope in the normal energy wavenumber spectrum as the spatial modulation factor of the dual-channel coupling kernel function;

[0027] Constructing a main kernel structure of a dual-channel coupling kernel function based on the temporal modulation factor and the spatial modulation factor, introducing the logarithmic decay rate of the tangential energy diffusivity as a dynamic attenuation factor of the dual-channel coupling kernel function, and generating a three-dimensional convolution kernel;

[0028] The initial value of the sliding radius is dynamically adjusted according to the meniscus curvature radius, and the direction vector of the convolution step is defined in combination with the anatomical characteristics of the ligament direction to generate dynamic window parameters that match the biomechanical characteristics of the knee joint.

[0029] Performing a sliding integral calculation in the biomechanical spatial domain using the three-dimensional convolution kernel and the dynamic window parameters, adjusting the attenuation factor weight of the three-dimensional convolution kernel according to the local energy density gradient during each sliding process, and generating an initial mapping of the interference intensity distribution matrix;

[0030] The initial mapping is subjected to local energy equalization processing, and the second-order derivative characteristics of the energy density gradient in the equalized mapping are extracted. The sliding radius and step size parameters of the three-dimensional convolution kernel are adjusted according to the second-order derivative characteristics to generate an interference intensity distribution matrix.

[0031] Optionally, the step of collecting the frequency characteristics of the patient's surface electromyographic signals and the tibial rotation angle during exercise, constructing an electromyographic activation pattern marked with a movement phase, and performing feature fusion on the biomechanical property map and the electromyographic activation pattern to generate a comprehensive profile of joint dynamics includes:

[0032] Perform multi-scale time-frequency decomposition on the surface electromyographic signals collected from patients during exercise, extract the energy concentration coefficient and phase synchronization index of specific frequency bands in the time-frequency domain, and generate electromyographic time-frequency feature vectors;

[0033] collecting time series data of tibial rotation angle, calculating the rotational dynamic stability coefficient of the time series data by the angle change rate, and constructing the electromyographic activation pattern marked by the movement stage in combination with the electromyographic time-frequency feature vector;

[0034] Performing spatial domain feature extraction on the biomechanical property atlas to obtain the cartilage contact stress distribution gradient and ligament tension change rate in the atlas, and generating a biomechanical feature vector;

[0035] Performing multimodal feature alignment on the electromyographic activation pattern and the biomechanical eigenvector, aligning the time series of the movement phases using a dynamic time warping algorithm, and generating a time-synchronized multimodal feature matrix;

[0036] A comprehensive joint dynamics profile is constructed based on the multimodal characteristic matrix, and the energy concentration coefficient in the myoelectric activation pattern is coupled with the stress distribution gradient in the biomechanical characteristic vector through a nonlinear mapping function to generate a comprehensive dynamics profile vector.

[0037] Optionally, the step of collecting biological parameters of the patient's lower limbs to construct a personalized knee joint model, performing vector superposition of the mechanical phase trajectory data and the six-degree-of-freedom contact force distribution data within the joint cavity to generate a dynamic mechanical fingerprint, and mapping the dynamic mechanical fingerprint into a biomechanical property map using an inverse dynamics solver includes:

[0038] Collecting the patient's lower limb length, interfemoral width, and tibial plateau inclination as lower limb biological parameters, and constructing a parametric knee joint mesh model based on the lower limb biological parameters, wherein the parametric knee joint mesh model includes the meniscus curvature radius and cruciate ligament attachment point coordinates that match the anatomical structure;

[0039] Performing trajectory smoothing processing on the mechanical phase trajectory data, extracting the patellar slip velocity vector and acceleration change rate in the mechanical phase trajectory, and generating a mechanical phase feature vector;

[0040] performing principal component analysis on the six-degree-of-freedom contact force distribution data within the joint cavity, extracting the normal force fluctuation amplitude and the tangential force coupling coefficient in the six-degree-of-freedom contact force distribution within the joint cavity, and generating a contact force eigenvector;

[0041] Performing vector superposition of the mechanical phase eigenvector and the contact force eigenvector, and generating a dynamic mechanical fingerprint through a weighted fusion algorithm, wherein the weighted fusion algorithm dynamically adjusts the weight ratio of the mechanical phase eigenvector and the contact force eigenvector according to the biological parameters of the lower limb;

[0042] The dynamic mechanical fingerprint is input into an inverse dynamics solver, and a mechanical inversion calculation is performed based on the parameterized knee joint mesh model to solve the cartilage contact stress distribution gradient and ligament tension change rate, thereby generating a biomechanical property map.

[0043] In a second aspect, an embodiment of the present application provides an artificial intelligence-assisted diagnosis system for a knee joint, comprising:

[0044] a receiving module, configured to receive a multi-angle dynamic X-ray image sequence and six-degree-of-freedom contact force distribution data within the joint cavity, dynamically track the multi-angle dynamic X-ray image sequence and, in combination with the six-degree-of-freedom contact force distribution data within the joint cavity, encode the patellar sliding trajectory to generate mechanical phase trajectory data;

[0045] a collection module for collecting biological parameters of the patient's lower limbs to construct a personalized knee joint model, performing vector superposition of the mechanical phase trajectory data with the six-degree-of-freedom contact force distribution data within the joint cavity to generate a dynamic mechanical fingerprint, and mapping the dynamic mechanical fingerprint into a biomechanical property map through an inverse dynamics solver;

[0046] A construction module is used to collect the frequency characteristics of the patient's surface electromyographic signals and the tibial rotation angle during exercise, construct an electromyographic activation pattern marked with the exercise phase, and perform feature fusion of the biomechanical property map with the electromyographic activation pattern to generate a comprehensive profile of joint dynamics;

[0047] An input module is configured to input the joint dynamics comprehensive profile into a pre-built joint abnormal dynamics propagation model, calculate the wave propagation path of the meniscus stress concentration area and the spatiotemporal evolution trajectory of the cruciate ligament deformation energy field using the stress wave conduction equation, and identify the compensatory injury pattern by analyzing the interference effect between the spatiotemporal evolution trajectory and the wave propagation path;

[0048] The adjustment module is used to generate a diagnostic map based on the compensatory injury pattern, combined with the biomechanical property map, the myoelectric activation pattern and the comprehensive profile of joint dynamics, match the biomechanical constraints in the preset motion feature library, and adaptively modify the diagnostic map to generate an individualized rehabilitation strategy.

[0049] In a third aspect, an embodiment of the present application provides a computing device comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute an artificial intelligence-assisted diagnosis method for a knee joint as described in any one of the first aspects.

[0050] In a fourth aspect, an embodiment of the present application provides a computer storage medium having computer program instructions stored thereon, which, when executed by a processor, implements an artificial intelligence-assisted diagnosis method for a knee joint as described in any one of the first aspects.

[0051] In an embodiment of the present application, a multi-angle dynamic X-ray image sequence and six-degree-of-freedom contact force distribution data in the joint cavity are received, the multi-angle dynamic X-ray image sequence is dynamically tracked and combined with the six-degree-of-freedom contact force distribution data in the joint cavity, the patellar slip trajectory is encoded to generate mechanical phase trajectory data; the patient's lower limb biological parameters are collected to construct a personalized knee joint model, the mechanical phase trajectory data and the six-degree-of-freedom contact force distribution data in the joint cavity are vector-superimposed to generate a dynamic mechanical fingerprint, and the dynamic mechanical fingerprint is mapped into a biomechanical property map through an inverse dynamics solver; the frequency characteristics of the patient's surface electromyographic signal and the tibial rotation angle during exercise are collected, the electromyographic activation pattern of the exercise stage marker is constructed, and the biological The mechanical property map is feature-fused with the electromyographic activation pattern to generate a comprehensive profile of joint dynamics; the comprehensive profile of joint dynamics is input into a pre-constructed joint abnormal dynamic propagation model, and the stress wave conduction equation is used to calculate the wave propagation path of the meniscus stress concentration area and the spatiotemporal evolution trajectory of the cruciate ligament deformation energy field, and the compensatory injury pattern is identified by analyzing the interference effect between the spatiotemporal evolution trajectory and the wave propagation path; according to the compensatory injury pattern, the biomechanical property map, the electromyographic activation pattern and the comprehensive profile of joint dynamics are combined to generate a diagnostic map, match the biomechanical constraints in the preset motion feature library, and adaptively modify the diagnostic map to generate an individualized rehabilitation strategy.

[0052] The technical solution of this application has the following beneficial effects:

[0053] This application not only improves the accuracy of complex structural injury pattern recognition of the knee joint, but also realizes the formulation of personalized rehabilitation strategies by integrating multi-source data (such as dynamic X-ray images, contact force distribution, electromyographic signals, etc.), effectively improving the treatment effect and patient recovery speed, but also provides a more scientific and systematic decision-making support tool for clinicians.

[0054] Furthermore, the embodiment of the present application also performs a hierarchical tensor decomposition on the comprehensive profile of joint dynamics to obtain biomechanical tensor components and kinematic tensor components. Based on the stress waveguide model, the meniscus stress propagation path is constructed for the biomechanical tensor components, and the dynamic correlation parameters of the fluctuation attenuation rate and the stress concentration coefficient are calculated; at the same time, the ligament energy field evolution model of the viscoelastic deformation field is established for the kinematic tensor components, and its spatiotemporal gradient and energy retention node coordinates are extracted. Through the above-mentioned dynamic correlation parameters, spatiotemporal gradient and energy retention node coordinates, the interference field equation is established by utilizing the dynamic coupling of the stress waveguide model and the viscoelastic deformation field, the wave field interference intensity distribution between the meniscus stress propagation path and the energy retention node is calculated, the abnormal resonance area and its three-dimensional coordinates and energy overflow threshold are identified, and the compensatory injury pattern vector is generated. Finally, the vector is modally matched with the standard motion chain parameters in the motion feature library, and the frequency domain weighted adjustment is performed based on the matching results to finally generate a compensatory injury pattern.

[0055] By integrating multi-source data and utilizing advanced mathematical models (such as stress waveguide models and viscoelastic deformation field models), this method can accurately calculate the spatiotemporal evolution of the meniscus stress propagation path and the cruciate ligament deformation energy field, thereby identifying compensatory injury patterns in the knee joint. This approach not only improves diagnostic accuracy and personalization, but also automatically generates rehabilitation strategies tailored to the patient's specific needs, significantly enhancing treatment outcomes and patient recovery speed. Furthermore, frequency-domain weighted adjustment further optimizes the accuracy of injury pattern recognition, ensuring high consistency between the diagnostic map and the actual clinical situation, providing strong support for precision medicine for knee joint diseases.

[0056] These and other aspects of the present application will become more readily apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0058] Figure 1 A flowchart of an artificial intelligence-assisted diagnosis method for a knee joint provided in an embodiment of the present application;

[0059] Figure 2 A schematic diagram of the structure of an artificial intelligence-assisted diagnosis system for a knee joint provided in an embodiment of the present application;

[0060] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0062] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to being different types.

[0063] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0064] Figure 1 A flowchart of an artificial intelligence-assisted diagnosis method for a knee joint is provided for an embodiment of the present application. Figure 1 As shown, the method includes:

[0065] Step 101: receiving a multi-angle dynamic X-ray image sequence and six-degree-of-freedom contact force distribution data within the joint cavity, dynamically tracking the multi-angle dynamic X-ray image sequence and combining it with the six-degree-of-freedom contact force distribution data within the joint cavity, encoding the patellar sliding trajectory to generate mechanical phase trajectory data;

[0066] In this step, a multi-angle dynamic X-ray imaging sequence is used. This is a medical imaging technique that captures continuous images of the internal structures of the human body from different angles to observe the movement of joints or other parts. These images can provide information on how bones and soft tissues interact. Six-degree-of-freedom contact force distribution data within the joint cavity refers to the magnitude and direction of the forces acting on the various parts of the joint during contact. Six degrees of freedom means that forces can move along three spatial dimensions (front and back, left and right, and up and down) and rotate about these three axes, fully reflecting the complex mechanical environment within the joint.

[0067] In practice, the system first synchronously collects multi-angle dynamic X-ray images of the patient and contact force distribution data within the joint cavity. Then, using image processing algorithms, it dynamically tracks the patellar sliding trajectory and combines this with the contact force data to encode phase trajectory data that reflects the mechanical properties of the knee joint. This process lays the foundation for the subsequent construction of a personalized knee joint model.

[0068] For example, an athlete presents with knee pain. The doctor first scans the athlete's knee using a multi-angle dynamic X-ray machine and simultaneously records the six-degree-of-freedom contact force distribution within the joint cavity. Based on this data, the system dynamically tracks the patella's sliding trajectory and discovers abnormal sliding patterns in certain movements. This provides critical information for developing a personalized knee model for the athlete.

[0069] Step 102: Collecting biological parameters of the patient's lower limbs to construct a personalized knee joint model, performing vector superposition of the mechanical phase trajectory data and the six-degree-of-freedom contact force distribution data within the joint cavity to generate a dynamic mechanical fingerprint, and mapping the dynamic mechanical fingerprint into a biomechanical property map using an inverse dynamics solver;

[0070] In this step, lower limb biological parameters, which include various physiological and anatomical information related to the individual's lower limbs, such as bone length, muscle mass, range of joint motion, etc., are used to build a knee joint model that accurately reflects individual differences. The personalized knee joint model is a computational model established based on the patient's specific biological parameters to simulate the function and potential problems of their knee joint. This model helps to understand the biomechanical characteristics of an individual in a healthy or injured state. The dynamic mechanical fingerprint is generated by superimposing the mechanical phase trajectory data and the contact force distribution data vector in the joint cavity. It represents the mechanical behavior pattern of the knee joint of a specific individual and can be used to identify abnormalities or disease characteristics.

[0071] In practice, the system constructs a knee joint model based on collected biometric parameters of the patient's lower limbs, accurately reflecting individual differences. It then uses mechanical phase trajectory data and contact force distribution data to generate a dynamic mechanical fingerprint. Through inverse dynamics analysis, this fingerprint is converted into a detailed biomechanical property map, displaying information such as cartilage stress and ligament tension.

[0072] For example, continuing with the above example, the doctor further collected the athlete's lower limb biometrics (such as bone length and muscle strength) and input them into the system to construct a personalized knee joint model. The system then used the previously acquired mechanical phase trajectory data and contact force distribution data to generate the athlete's dynamic mechanical fingerprint. Through inverse dynamics analysis, it generated a biomechanical property map, revealing the specific damage to the knee joint.

[0073] Step 103: collecting the time-frequency characteristics of the patient's surface electromyographic signal and the tibial rotation angle during the movement, constructing an electromyographic activation pattern marked with the movement phase, and performing feature fusion on the biomechanical property map and the electromyographic activation pattern to generate a comprehensive profile of joint dynamics;

[0074] In this step, the surface EMG signal's time-frequency characteristics are used to assess muscle activity by recording changes in muscle electrical activity. Time-frequency characteristics refer to the signal's characteristics over time and frequency, reflecting the degree of muscle activation and coordination during movement. Tibial rotation refers to the angle of rotation of the shin bone (tibia) relative to the thigh bone (femur). It is crucial for analyzing gait and knee stability during movement. The EMG activation pattern is a model constructed from the collected surface EMG signals and other relevant data to demonstrate the activity patterns of various muscle groups during specific movement phases.

[0075] In practice, the system collects surface electromyographic signals and tibial rotation angle data during patient movement, constructs an electromyographic activation pattern, and fuses it with the previously generated biomechanical property map to form a comprehensive profile vector that fully describes the functional state of the knee joint. This step helps to more accurately identify complex injury patterns in the knee joint.

[0076] For example, in the next step, the doctor had the athlete perform several standard motion tests while simultaneously recording their surface electromyographic signals and tibial rotation angle. The system constructed the athlete's electromyographic activation pattern from this data and, by combining it with the previously generated biomechanical property map, generated a detailed, comprehensive profile of the joint dynamics, showing how the knee performed during movement.

[0077] Step 104: Inputting the joint dynamics comprehensive profile into a pre-built joint abnormal dynamics propagation model, using the stress wave conduction equation to calculate the wave propagation path of the meniscus stress concentration area and the spatiotemporal evolution trajectory of the cruciate ligament deformation energy field, and identifying the compensatory injury pattern by analyzing the interference effect between the spatiotemporal evolution trajectory and the wave propagation path;

[0078] In this step, the stress wave propagation equation is a mathematical formula used to describe the propagation of stress waves within solid materials due to external forces. In this context, it is used to simulate the response of the meniscus and cruciate ligaments to external forces. The wave propagation path is the direction and path of the stress wave propagating within the tissue when the force applied to the meniscus causes it to propagate. This is crucial for understanding the injury mechanism. The spatiotemporal evolution trajectory of the deformation energy field describes how the energy distribution of the cruciate ligament changes over time due to the external force, which helps analyze the development of the injury.

[0079] In practice, the system inputs the comprehensive joint dynamics profile generated in step 103 into a pre-built model to simulate the propagation path of stress waves within the knee joint and the energy field changes of the cruciate ligaments. By analyzing this data, the system can identify key areas and patterns that may lead to compensatory injuries.

[0080] For example, the system input a previously generated comprehensive joint dynamics profile into the model, simulating the propagation path of stress waves within the athlete's knee joint and identifying areas of stress concentration in the meniscus during certain motions. Furthermore, the system analyzed changes in the energy field of the cruciate ligaments and identified several key locations that could potentially trigger compensatory injuries, providing a basis for further diagnosis.

[0081] Step 105: Based on the compensatory injury pattern, combined with the biomechanical property map, the myoelectric activation pattern and the comprehensive profile of joint dynamics, a diagnostic map is generated, the biomechanical constraints in the preset motion feature library are matched, and the diagnostic map is adaptively modified to generate an individualized rehabilitation strategy.

[0082] In this step, compensatory injury patterns are defined as adjustments made by other parts of the body to compensate for the loss of function after injury, potentially leading to new injuries. This pattern helps identify these secondary injuries. The diagnostic profile integrates all previous data analysis results into a comprehensive report to guide treatment decisions. It can include information such as biomechanical properties, myoelectric activity patterns, and kinetic profile vectors. A personalized rehabilitation strategy is a recovery plan tailored to the patient's specific circumstances, designed to optimize rehabilitation outcomes and prevent future injuries. This strategy is customized based on an understanding of the patient's specific biomechanical constraints.

[0083] In practice, the system integrates all previous data and analysis results to generate a detailed diagnostic profile, which is then adjusted based on a pre-set library of movement characteristics. Ultimately, the system outputs a personalized rehabilitation strategy for the patient, ensuring the scientific and effective treatment plan.

[0084] For example, the system generated a detailed diagnostic map based on all previous data, illustrating the athlete's knee injury and its compensation mechanism. The system also matched the relevant constraints in the motion feature library, adaptively revising the diagnostic map and ultimately developing a personalized rehabilitation training plan for the athlete, helping them recover faster.

[0085] This method integrates multi-source data (such as dynamic X-ray images, contact force distribution, and electromyographic signals) and utilizes advanced mathematical models to accurately identify complex knee injury patterns and automatically generate personalized rehabilitation strategies. This approach not only improves diagnostic accuracy and personalization, but also effectively guides clinical treatment, significantly improving patients' recovery speed and quality of life. Each step is closely linked, gradually and deeply analyzing the biomechanical properties of the knee joint, providing strong support for precision medicine.

[0086] To address the problem of insufficient accuracy of existing methods in identifying compensatory knee injury patterns, in certain embodiments, step 104 includes inputting the comprehensive joint dynamics profile into a pre-built joint abnormal dynamics propagation model, using the stress wave conduction equation to calculate the wave propagation path of the meniscus stress concentration area and the spatiotemporal evolution trajectory of the cruciate ligament deformation energy field, and identifying the compensatory injury pattern by analyzing the interference effect between the spatiotemporal evolution trajectory and the wave propagation path, including:

[0087] The joint dynamics comprehensive profile is subjected to hierarchical tensor decomposition to obtain biomechanical tensor components and kinematic tensor components. A meniscus stress propagation path based on a stress waveguide model is constructed for the biomechanical tensor components, and dynamic correlation parameters between the fluctuation attenuation rate and the stress concentration factor in the meniscus stress propagation path are calculated. A ligament energy field evolution model of the viscoelastic deformation field is established for the kinematic tensor components, and the spatiotemporal gradient and energy retention node coordinates of the ligament energy field evolution model are extracted. An interference field equation is established through dynamic coupling of the stress waveguide model and the viscoelastic deformation field using the dynamic correlation parameters, the spatiotemporal gradient, and the energy retention node coordinates. A wavefield interference intensity distribution between the meniscus stress propagation path and the energy retention node coordinates is calculated based on the interference field equation. A compensatory injury pattern vector is generated based on the three-dimensional coordinates of the abnormal resonance region in the wavefield interference intensity distribution and the energy overflow threshold. The compensatory injury pattern vector is modally matched with standard kinematic chain parameters in a motion feature library, and the compensatory injury pattern vector is subjected to frequency domain weighted adjustment based on the matching results to generate a compensatory injury pattern.

[0088] In this embodiment, hierarchical tensor decomposition is a data processing technique used to decompose complex multidimensional data (such as a comprehensive joint dynamics profile) into multiple, more easily analyzable components. Specifically, in this application, the comprehensive joint dynamics profile is decomposed into biomechanical and kinematic tensor components. The biomechanical tensor component primarily contains information about the mechanical properties of the internal structures of the knee joint (such as the meniscus and ligaments), such as stress distribution and deformation. This data is used to construct the meniscus stress propagation path based on the stress waveguide model and calculate the dynamic correlation parameters between its fluctuation attenuation rate and stress concentration factor. The kinematic tensor component focuses on the geometric and temporal characteristics of the knee joint during motion, such as tibial rotation angle and muscle activation pattern. This data is used to establish a ligament energy field evolution model based on the viscoelastic deformation field and extract the spatiotemporal gradient of the energy field and the coordinates of the energy retention nodes. The interference field equation is a mathematical model established based on the dynamic coupling of the stress waveguide model and the viscoelastic deformation field, used to calculate the wave field interference intensity distribution between the meniscus stress propagation path and the energy retention nodes. By analyzing the abnormal resonance region in the distribution and its three-dimensional coordinates and energy overflow threshold, a compensatory damage pattern vector is generated.

[0089] In the embodiment of the present application, first, the comprehensive profile of joint dynamics is subjected to hierarchical tensor decomposition to obtain biomechanical tensor components and kinematic tensor components. Then, the biomechanical tensor components are used to construct the meniscus stress propagation path based on the stress waveguide model, and the dynamic correlation parameters of the fluctuation attenuation rate and the stress concentration coefficient are calculated. At the same time, the kinematic tensor components are used to establish the ligament energy field evolution model of the viscoelastic deformation field, and its spatiotemporal gradient and energy retention node coordinates are extracted. Then, through the above-mentioned dynamic correlation parameters, spatiotemporal gradient and energy retention node coordinates, the interference field equation is established by the dynamic coupling of the stress waveguide model and the viscoelastic deformation field, and the wavefield interference intensity distribution is calculated. Finally, the compensatory injury pattern vector is generated according to the abnormal resonance area in the wavefield interference intensity distribution and its three-dimensional coordinates and the energy overflow threshold, and it is modally matched with the standard motion chain parameters in the motion feature library, and the frequency domain weighted adjustment is performed based on the matching results to generate the final compensatory injury pattern.

[0090] The following is a specific embodiment:

[0091] A patient presented with recurring knee pain. The physician first collected imaging and mechanical data from the patient's knee using a multi-angle dynamic X-ray machine and a six-degree-of-freedom contact force sensor. This data was then combined with surface electromyography (EMG) signals and tibial rotation angle data to generate a comprehensive profile of joint dynamics. The system then performed a hierarchical tensor decomposition of this comprehensive profile, yielding biomechanical and kinematic tensor components. In the biomechanical tensor component, the system constructed a stress waveguide model for the meniscus stress propagation path and calculated the dynamic correlation parameters between the fluctuation attenuation rate and the stress concentration coefficient. The results showed significant stress concentration in the meniscus under certain specific movements, indicating a potential injury risk. Furthermore, in the kinematic tensor component, the system established a ligament energy field evolution model based on the viscoelastic deformation field, extracting its spatiotemporal gradients and the coordinates of energy retention nodes. This analysis revealed abnormal fluctuations in the cruciate ligament energy field during walking, suggesting the possibility of early injury or dysfunction. Subsequently, the system established the interference field equation by dynamically correlating parameters, spatiotemporal gradients, and energy retention node coordinates, using the dynamic coupling of the stress waveguide model and the viscoelastic deformation field, and calculated the wave field interference intensity distribution. The results showed that there was an obvious abnormal resonance area between the meniscus stress propagation path and the energy retention node, further confirming the presence of a compensatory injury pattern in the patient's knee joint. Finally, the system generated a compensatory injury pattern vector and performed modal matching with the standard motion chain parameters in the motion feature library. Based on the matching results, it performed frequency domain weighted adjustment to generate a personalized rehabilitation strategy. This strategy not only helped doctors accurately diagnose patients' knee joint problems, but also provided a scientific basis for formulating effective treatment plans. This method significantly improved the accuracy and personalization of diagnosis, helping to improve patients' recovery speed and quality of life.

[0092] To further improve the accuracy and personalization of knee joint compensatory injury pattern recognition, in certain embodiments, step 203 establishes an interference field equation through the dynamic correlation parameters, the spatiotemporal gradient, and the energy residence node coordinates, by dynamically coupling a stress waveguide model with a viscoelastic deformation field; calculates the wavefield interference intensity distribution between the meniscus stress propagation path and the energy residence node coordinates according to the interference field equation; and generates a compensatory injury pattern vector according to the three-dimensional coordinates of the abnormal resonance region in the wavefield interference intensity distribution and the energy overflow threshold, including:

[0093] The meniscus stress propagation path is parameterized and reconstructed using a wave equation, and the wave phase delay parameter and amplitude decay rate in the meniscus stress propagation path are extracted. The energy residence node coordinates are decomposed into an energy gradient field to generate an energy density gradient vector and a residence time coefficient. A dynamic coupling kernel function is constructed based on the wave phase delay parameter and the energy density gradient vector. The dynamic coupling kernel function is combined with a spatial convolution operation to generate a spatial grid map of the wave field interference intensity distribution, and grid cells with energy density peaks higher than an overflow threshold are marked in the spatial grid map as abnormal resonance regions. The amplitude decay rate and the residence time coefficient are used to calculate the product of the energy residence time coefficient and the wave amplitude decay rate in the abnormal resonance region as a compensatory damage intensity factor, and the angle between the three-dimensional coordinates of the abnormal resonance region and the tangent direction of the wave propagation path is extracted as a damage spatial distribution parameter. The compensatory damage intensity factor and the damage spatial distribution parameter are tensor-recombined to generate a compensatory damage pattern vector.

[0094] In this embodiment, wave equation parameterization reconstruction is a process of mathematically modeling the meniscus stress propagation path. The propagation characteristics of stress waves in tissue are described by extracting the wave phase delay parameter and amplitude decay rate. These parameters help understand the propagation behavior of stress waves in different media. Energy gradient field decomposition is a technique that decomposes the coordinates of energy-residing nodes into energy density gradient vectors and dwell time coefficients. The energy density gradient vector describes the spatial variation trend of energy, while the dwell time coefficient reflects the length of time that energy resides at a specific location. A dynamic coupling kernel function, based on the wave phase delay parameter and energy density gradient vector, constructs a mathematical model to describe the dynamic coupling relationship between the stress waveguide model and the viscoelastic deformation field. This kernel function, combined with spatial convolution operations, generates a spatial grid mapping of the wavefield interference intensity distribution. The compensatory damage intensity factor is calculated by multiplying the energy dwell time coefficient and the wave amplitude decay rate in the abnormal resonance region and is used to quantify the severity of the damage. Furthermore, by extracting the angle between the three-dimensional coordinates of the abnormal resonance region and the tangent direction of the wave propagation path as a parameter of the spatial damage distribution, the specific location and morphology of the damage can be further refined.

[0095] In an embodiment of the present application, first, the wave equation parameterized reconstruction of the meniscus stress propagation path is performed, and the wave phase delay parameter and amplitude decay rate are extracted. Then, the energy gradient field is decomposed on the energy residence node coordinates to generate an energy density gradient vector and a residence time coefficient. A dynamic coupling kernel function is constructed based on the wave phase delay parameter and the energy density gradient vector, and the kernel function is combined with a spatial convolution operation to generate a spatial grid mapping of the wave field interference intensity distribution. In this mapping, the grid cells with energy density peaks higher than the overflow threshold are marked as abnormal resonance areas. Then, the amplitude decay rate and the residence time coefficient are used to calculate the compensatory damage intensity factor in the abnormal resonance area, and the angle between the three-dimensional coordinates of the abnormal resonance area and the tangent direction of the wave propagation path is extracted as the damage space distribution parameter. Finally, the compensatory damage intensity factor and the damage space distribution parameter are tensor-reorganized to generate a compensatory damage pattern vector.

[0096] The following is a specific embodiment:

[0097] An athlete presented with recurring knee pain. The physician first collected knee imaging and mechanical data using a multi-angle dynamic X-ray machine and a six-degree-of-freedom contact force sensor. This data was combined with surface electromyography (EMG) signals and tibial rotation angle data to generate a comprehensive profile of the joint's dynamics. The system then analyzed this comprehensive profile, focusing specifically on the meniscus stress propagation paths and energy retention node coordinates. Within the biomechanical tensor component, the system performed a wave equation parameterized reconstruction of the meniscus stress propagation paths, extracting the phase delay parameters and amplitude decay rates. The results revealed significant stress concentration in the meniscus under certain specific movements, indicating a potential injury risk. Furthermore, within the kinematic tensor component, the system performed an energy gradient field decomposition of the energy retention node coordinates, generating energy density gradient vectors and dwell time coefficients. This analysis revealed abnormal fluctuations in the cruciate ligament energy field during walking, suggesting the possibility of early injury or dysfunction. The system then constructed a dynamic coupling kernel function based on the phase delay parameters and energy density gradient vectors. This kernel function, combined with a spatial convolution operation, generated a spatial grid map of the wavefield interference intensity distribution. In this mapping, grid cells with energy density peaks above the overflow threshold are marked as abnormal resonance areas. The results show that there is a clear abnormal resonance area between the meniscus stress propagation path and the energy retention node, further confirming the presence of a compensatory injury pattern in the patient's knee joint. Next, the system uses the amplitude decay rate and the residence time coefficient to calculate the compensatory injury intensity factor in the abnormal resonance area, and extracts the angle between the three-dimensional coordinates of the abnormal resonance area and the tangent direction of the wave propagation path as the injury space distribution parameter. Finally, the system performs tensor reorganization on the compensatory injury intensity factor and the injury space distribution parameter to generate a compensatory injury pattern vector. This detailed analysis not only helps doctors accurately diagnose patients' knee joint problems, but also provides information about the severity and specific location of the injury, providing a scientific basis for formulating an effective treatment plan. This method significantly improves the accuracy and personalization of diagnosis, helping to improve patients' recovery speed and quality of life.

[0098] In order to further improve the accuracy of pattern recognition and spatial positioning of knee compensatory injury, in certain embodiments, step 302 constructs a dynamic coupling kernel function based on the fluctuation phase delay parameter and the energy density gradient vector, generates a spatial grid map of the wavefield interference intensity distribution by combining the dynamic coupling kernel function with a spatial convolution operation, and marks grid cells in the spatial grid map with energy density peaks exceeding an overflow threshold as abnormal resonance regions, including:

[0099] The fluctuation phase delay parameter is subjected to time series normalization processing to generate a phase delay weight coefficient, and the energy density gradient vector is subjected to orthogonal basis decomposition to obtain a normal energy gradient component and a tangential energy diffusivity. A dual-channel coupling kernel function is constructed based on the phase delay weight coefficient and the normal energy gradient component, and the tangential energy diffusivity is used as a dynamic attenuation factor of the dual-channel coupling kernel function. According to the dynamic attenuation factor, a three-dimensional convolution kernel is generated, and the dynamic window parameters of the spatial convolution operation are defined according to the anatomical structure of the knee joint. The dynamic window parameters are combined with the three-dimensional convolution kernel to perform a sliding integral calculation in the biomechanical space domain to generate an interference intensity distribution matrix. The interference intensity distribution matrix is ​​subjected to multi-scale Gaussian filtering processing, and the spatial topological structure of the energy density peak in the filtered matrix is ​​detected. The topological units that are continuously adjacent in the spatial topological structure and whose peak values ​​are higher than the overflow threshold are clustered as abnormal resonance regions.

[0100] In this embodiment, time series normalization processing is the process of standardizing the fluctuation phase delay parameters so that they can be compared between different time points. The phase delay weight coefficient generated by this processing can better reflect the propagation characteristics of stress waves at different time points. Orthogonal basis decomposition is a technique for decomposing the energy density gradient vector into multiple independent components. Specifically in this application, it decomposes the energy density gradient vector into a normal energy gradient component and a tangential energy diffusivity, which describe the energy variation trend in the directions perpendicular and parallel to the surface, respectively. The dual-channel coupling kernel function is a mathematical model constructed based on the phase delay weight coefficient and the normal energy gradient component, which is used to describe the coupling relationship between the stress waveguide model and the viscoelastic deformation field. The tangential energy diffusivity acts as a dynamic attenuation factor, allowing the kernel function to adjust its response according to actual conditions. The three-dimensional convolution kernel is a multidimensional filter generated based on the dual-channel coupling kernel function, which is used to perform sliding integral calculations in the biomechanical spatial domain. Combined with the dynamic window parameters of the spatial convolution operation defined by the anatomical structure of the knee joint, the propagation path of the stress wave in the tissue can be more accurately captured. Multi-scale Gaussian filtering is an image processing technique used to smooth data and reduce noise. By detecting the spatial topological structure of the energy density peaks in the filtered matrix, we can identify consecutively adjacent topological units with peaks above the overflow threshold, and these units are clustered into anomalous resonance regions.

[0101] In an embodiment of the present application, first, the fluctuation phase delay parameter is normalized in time series to generate a phase delay weight coefficient; at the same time, the energy density gradient vector is decomposed into an orthogonal basis to obtain the normal energy gradient component and the tangential energy diffusivity. Then, a dual-channel coupling kernel function is constructed based on the phase delay weight coefficient and the normal energy gradient component, and the tangential energy diffusivity is used as a dynamic attenuation factor to generate a three-dimensional convolution kernel. The dynamic window parameters of the spatial convolution operation are defined according to the anatomical structure of the knee joint, and the parameters are combined with the three-dimensional convolution kernel to perform a sliding integral calculation in the biomechanical space domain to generate an interference intensity distribution matrix. Finally, the interference intensity distribution matrix is ​​subjected to multi-scale Gaussian filtering, and the spatial topological structure of the energy density peak in the filtered matrix is ​​detected, and the topological units that are continuously adjacent and whose peak values ​​are higher than the overflow threshold are clustered as abnormal resonance regions.

[0102] The following is a specific embodiment:

[0103] A patient presented with recurring knee pain. The physician first collected imaging and mechanical data of the patient's knee using a multi-angle dynamic X-ray machine and a six-degree-of-freedom contact force sensor. This data was then combined with surface electromyography (EMG) signals and tibial rotation angle data to generate a comprehensive profile of the joint's dynamics. The system then conducted an in-depth analysis of this comprehensive profile, focusing specifically on the stress propagation paths and energy retention node coordinates within the meniscus. During this analysis, the system first normalized the fluctuation phase delay parameters over time to generate phase delay weight coefficients. The results revealed significant stress concentration in the meniscus under certain specific movements, indicating a potential injury risk. The system also performed an orthogonal basis decomposition of the energy density gradient vector to obtain the normal energy gradient component and the tangential energy diffusivity. This analysis revealed abnormal fluctuations in the cruciate ligament energy field during walking, suggesting the possibility of early injury or dysfunction. The system then constructed a dual-channel coupled kernel function based on the phase delay weight coefficients and the normal energy gradient component, generating a three-dimensional convolution kernel using the tangential energy diffusivity as a dynamic attenuation factor. The dynamic window parameters of the spatial convolution operation were defined based on the anatomical structure of the knee joint. These parameters, combined with the three-dimensional convolution kernel, were used to perform a sliding integral calculation in the biomechanical spatial domain, generating an interference intensity distribution matrix. The system then applied multi-scale Gaussian filtering to the interference intensity distribution matrix and detected the spatial topological structure of the energy density peaks in the filtered matrix. The results showed the presence of distinct abnormal resonance regions between the meniscus stress propagation path and the energy retention nodes, further confirming the presence of a compensatory injury pattern in the patient's knee joint. Ultimately, the system clustered consecutive, adjacent topological units with peak values ​​above the overflow threshold into abnormal resonance regions, providing detailed information on the location and morphology of the injury. This detailed analysis not only helped doctors accurately diagnose patients' knee joint problems but also provided a scientific basis for developing effective treatment plans. This approach significantly improved the accuracy and personalization of diagnosis, helping to improve patients' recovery speed and quality of life.

[0104] In order to improve the accuracy and spatial resolution of knee injury pattern recognition, in certain embodiments, step 402 constructs a dual-channel coupling kernel function based on the phase delay weight coefficient and the normal energy gradient component, uses the tangential energy diffusivity as a dynamic attenuation factor of the dual-channel coupling kernel function, generates a three-dimensional convolution kernel based on the dynamic attenuation factor, defines dynamic window parameters of the spatial convolution operation based on the anatomical structure of the knee joint, and performs a sliding integral calculation in the biomechanical spatial domain using the dynamic window parameters in combination with the three-dimensional convolution kernel to generate an interference intensity distribution matrix, including:

[0105] The phase delay weight coefficient is transformed in the frequency domain to generate a phase delay spectrum, the main frequency band energy ratio in the phase delay spectrum is extracted as the time modulation factor of the dual-channel coupling kernel function, the normal energy gradient component is transformed in the space Fourier transform to generate a normal energy wavenumber spectrum, and the high-frequency attenuation slope in the normal energy wavenumber spectrum is extracted as the spatial modulation factor of the dual-channel coupling kernel function; the main kernel structure of the dual-channel coupling kernel function is constructed based on the time modulation factor and the spatial modulation factor, the logarithmic attenuation rate of the tangential energy diffusivity is introduced as the dynamic attenuation factor of the dual-channel coupling kernel function, and a three-dimensional convolution kernel is generated; the sliding is dynamically adjusted according to the curvature radius of the meniscus The initial value of the radius is determined, and the direction vector of the convolution step is defined in combination with the anatomical characteristics of the ligament direction, so as to generate dynamic window parameters that match the biomechanical characteristics of the knee joint; a sliding integral calculation is performed in the biomechanical space domain using the three-dimensional convolution kernel and the dynamic window parameters, and the attenuation factor weight of the three-dimensional convolution kernel is adjusted according to the local energy density gradient during each sliding process to generate an initial mapping of the interference intensity distribution matrix; the initial mapping is subjected to local energy equalization processing, and the second-order derivative characteristics of the energy density gradient in the equalized mapping are extracted. The sliding radius and step parameters of the three-dimensional convolution kernel are adjusted according to the second-order derivative characteristics to generate an interference intensity distribution matrix.

[0106] In this embodiment, the frequency domain transform is performed on the phase delay weight coefficient to generate a phase delay spectrum, from which the main frequency band energy ratio is extracted as the temporal modulation factor of the dual-channel coupling kernel function. This step helps capture the propagation characteristics of stress waves at different frequencies. The spatial Fourier transform is performed on the normal energy gradient component to generate a normal energy wavenumber spectrum, from which the high-frequency attenuation slope is extracted as the spatial modulation factor of the dual-channel coupling kernel function. This step is used to analyze the spatial distribution characteristics of energy. The dynamic attenuation factor introduces the logarithmic decay rate of the tangential energy diffusivity as the dynamic attenuation factor of the dual-channel coupling kernel function, allowing the kernel function to adjust according to the actual energy diffusion situation, thereby more accurately simulating the propagation path of stress waves. The dynamic window parameter dynamically adjusts the initial value of the sliding radius based on the meniscus curvature radius and defines the direction vector of the convolution step size based on the anatomical characteristics of the ligament orientation, generating dynamic window parameters that match the biomechanical characteristics of the knee joint. These parameters ensure that the convolution kernel can perform effective sliding integral calculations on specific anatomical structures. The local energy equalization processing is to perform local energy equalization on the initial mapping, extract the second-order derivative features of the energy density gradient in the equalized mapping, and adjust the sliding radius and step size parameters of the three-dimensional convolution kernel through these features to generate the final interference intensity distribution matrix.

[0107] In this embodiment, the phase delay weight coefficient is first transformed in the frequency domain to generate a phase delay spectrum, and the main frequency band energy fraction is extracted as the temporal modulation factor. Simultaneously, the normal energy gradient component is transformed in the spatial Fourier transform to generate a normal energy wavenumber spectrum, and the high-frequency attenuation slope is extracted as the spatial modulation factor. Next, a dual-channel coupling kernel function is constructed based on the temporal and spatial modulation factors, and the logarithmic decay rate of the tangential energy diffusivity is introduced as a dynamic attenuation factor to generate a three-dimensional convolution kernel. The initial sliding radius is dynamically adjusted based on the meniscus curvature radius, and the direction vector of the convolution step is defined based on the anatomical characteristics of the ligament orientation to generate dynamic window parameters. Then, a sliding integral calculation is performed in the biomechanical spatial domain using the three-dimensional convolution kernel and the dynamic window parameters. During each sliding process, the attenuation factor weight of the three-dimensional convolution kernel is adjusted based on the local energy density gradient to generate an initial mapping of the interference intensity distribution matrix. Finally, the initial mapping is subjected to local energy equalization, and the second-order derivative characteristics of the energy density gradient in the equalized mapping are extracted. These characteristics are used to adjust the sliding radius and step size parameters of the three-dimensional convolution kernel to generate the final interference intensity distribution matrix.

[0108] The following is a specific embodiment:

[0109] A patient presented with recurring knee pain. Using a multi-angle dynamic X-ray machine and a six-degree-of-freedom contact force sensor, the doctor collected imaging and mechanical data from the patient's knee joint. This data was then combined with surface electromyography (EMG) signals and tibial rotation angle data to generate a comprehensive profile of joint dynamics. The system first performed a frequency domain transform on the fluctuation phase delay parameters to generate a phase delay spectrum and extracted the energy fraction of the main frequency band as a temporal modulation factor. The results revealed significant stress concentration in the meniscus under certain specific movements, indicating a potential injury risk. The system also performed a spatial Fourier transform on the normal energy gradient component to generate a normal energy wavenumber spectrum and extracted the high-frequency attenuation slope as a spatial modulation factor. This analysis revealed abnormal fluctuations in the cruciate ligament energy field during walking, suggesting the possibility of early injury or dysfunction. The system then constructed a dual-channel coupling kernel function based on the temporal and spatial modulation factors. The system also introduced the logarithmic decay rate of the tangential energy diffusivity as a dynamic attenuation factor to generate a three-dimensional convolution kernel. The initial value of the sliding radius is dynamically adjusted based on the radius of curvature of the meniscus, and the direction vector of the convolution step is defined in combination with the anatomical characteristics of the ligament orientation, generating dynamic window parameters that match the biomechanical characteristics of the knee joint. A sliding integral calculation is performed in the biomechanical spatial domain using the three-dimensional convolution kernel and dynamic window parameters. During each sliding process, the attenuation factor weight of the three-dimensional convolution kernel is adjusted based on the local energy density gradient, generating an initial mapping of the interference intensity distribution matrix. Finally, the system performs local energy equalization on the initial mapping and extracts the second-order derivative features of the energy density gradient in the equalized mapping. These features are used to adjust the sliding radius and step parameters of the three-dimensional convolution kernel to generate the final interference intensity distribution matrix. This provides detailed information on the location and morphology of the injury, helping doctors accurately diagnose patients' knee joint problems and providing a scientific basis for formulating effective treatment plans. This significantly improves the accuracy and personalization of diagnosis, contributing to improved patients' recovery speed and quality of life.

[0110] To further improve the accuracy of knee injury pattern recognition and the effectiveness of multimodal data fusion, in certain embodiments, the step 103 includes collecting the frequency characteristics of the patient's surface electromyographic signals and the tibial rotation angle during exercise, constructing an electromyographic activation pattern marked with the exercise phase, and performing feature fusion of the biomechanical property map with the electromyographic activation pattern to generate a comprehensive profile of joint dynamics, including:

[0111] The surface electromyographic signals collected from the patient during exercise are subjected to multi-scale time-frequency decomposition, and the energy concentration coefficient and phase synchronization index of a specific frequency band in the time-frequency domain are extracted to generate an electromyographic time-frequency feature vector; the time series data of the tibial rotation angle are collected, and the rotational dynamic stability coefficient of the time series data is calculated by the angle change rate, and the electromyographic activation pattern marked by the movement stage is constructed in combination with the electromyographic time-frequency feature vector; the spatial domain feature extraction is performed on the biomechanical property atlas, and the cartilage contact stress distribution gradient and ligament tension change rate in the atlas are obtained to generate a biomechanical feature vector; the electromyographic activation pattern and the biomechanical feature vector are multimodally aligned, and the time series of the movement stage are aligned by a dynamic time warping algorithm to generate a time-synchronized multimodal feature matrix; a comprehensive joint dynamics profile is constructed based on the multimodal feature matrix, and the energy concentration coefficient in the electromyographic activation pattern is coupled with the stress distribution gradient in the biomechanical feature vector by a nonlinear mapping function to generate a dynamic comprehensive profile vector.

[0112] In this embodiment, multiscale time-frequency decomposition is a signal processing technique used to extract features at different time scales and frequency ranges from surface electromyographic (EMG) signals. This method generates an EMG time-frequency feature vector containing the energy concentration coefficient and phase synchronization index for specific frequency bands. These features help describe muscle activation during different phases of movement. The rotational dynamic stability coefficient is evaluated by analyzing the collected time series data of tibial rotation angle and calculating its rate of change. This coefficient reflects the stability and coordination of the knee joint during movement. The cartilage contact stress distribution gradient and ligament tension change rate are spatial domain features extracted from the biomechanical property map and are used to describe the mechanical response of internal knee structures (such as cartilage and ligaments) under different motion states. These features provide important information about knee joint health. Multimodal feature alignment uses the dynamic time warping (DTW) algorithm to align data from different modalities (such as EMG signals and biomechanical features) onto the same time axis, ensuring temporal synchronization of the modal data and generating a time-synchronized multimodal feature matrix. A nonlinear mapping function is a mathematical model that couples the energy concentration coefficients in the myoelectric activation pattern with the stress distribution gradients in the biomechanical eigenvectors to generate a comprehensive dynamic profile vector. This coupling helps capture complex joint dynamics.

[0113] In an embodiment of the present application, first, the surface electromyographic signals collected from the patient during exercise are subjected to multi-scale time-frequency decomposition, the energy concentration coefficient and phase synchronization index of a specific frequency band in the time-frequency domain are extracted, and an electromyographic time-frequency feature vector is generated. At the same time, the time series data of the tibial rotation angle are collected, the rotational dynamic stability coefficient is calculated by the angle change rate, and the electromyographic activation pattern marked by the movement stage is constructed in combination with the electromyographic time-frequency feature vector. Next, the biomechanical property map is subjected to spatial domain feature extraction to obtain the cartilage contact stress distribution gradient and the ligament tension change rate to generate a biomechanical feature vector. Then, the electromyographic activation pattern is multimodally aligned with the biomechanical feature vector, and the time series of the movement stage is aligned by a dynamic time warping algorithm to generate a time-synchronized multimodal feature matrix. Finally, a comprehensive profile of joint dynamics is constructed based on the multimodal feature matrix, and the energy concentration coefficient in the electromyographic activation pattern is coupled with the stress distribution gradient in the biomechanical feature vector by a nonlinear mapping function to generate a comprehensive dynamic profile.

[0114] The following is a specific embodiment:

[0115] An athlete presented with recurring knee pain. The physician first collected imaging and mechanical data of the patient's knee using a multi-angle dynamic X-ray machine and a six-degree-of-freedom contact force sensor. This data was then combined with surface electromyography (EMG) signals and tibial rotation angle data to generate a comprehensive profile of joint dynamics. The system then performed a multiscale time-frequency decomposition of the surface electromyography (EMG) signals, extracting energy concentration coefficients and phase synchronization indices for specific frequency bands and generating EMG time-frequency feature vectors. The results revealed abnormal muscle activation patterns during certain movements, indicating potential muscle fatigue or injury. The system also collected time series data on tibial rotation angles and calculated the rate of change of angle to determine the rotational dynamic stability coefficient. The system found that the patient's knee rotational stability was poor during walking, suggesting the possibility of early injury or dysfunction. The system then performed spatial feature extraction on the biomechanical property map, obtaining the cartilage contact stress gradient and ligament tension change rate, and generated a biomechanical feature vector. This analysis revealed that the patient experienced elevated stress on the meniscus and cruciate ligaments during certain movements, potentially contributing to the pain. The system then performed multimodal feature alignment between the EMG activation patterns and the biomechanical eigenvectors. Using a dynamic time warping algorithm, the system aligned the time series of the movement phases, generating a time-synchronized multimodal feature matrix. This process ensured precise temporal correspondence between the EMG signals and the biomechanical features, improving the accuracy of data analysis. Finally, the system constructed a comprehensive joint dynamics profile based on the multimodal feature matrix. Using a nonlinear mapping function, the system coupled the energy concentration coefficient in the EMG activation patterns with the stress distribution gradient in the biomechanical eigenvectors to generate a comprehensive dynamics profile vector. This detailed analysis not only helped doctors accurately diagnose patients' knee problems but also provided a scientific basis for developing effective treatment plans. This approach significantly improved the accuracy and personalization of diagnoses, contributing to improved recovery and quality of life for patients. The entire process demonstrated how multimodal data fusion technology can be used to enhance the comprehensiveness and accuracy of knee injury identification.

[0116] To further improve the accuracy of constructing a personalized knee joint model and the effectiveness of generating a biomechanical property map, in certain embodiments, the step 102 involves collecting biological parameters of the patient's lower limb to construct a personalized knee joint model, performing vector superposition of the mechanical phase trajectory data with the six-degree-of-freedom contact force distribution data within the joint cavity to generate a dynamic mechanical fingerprint, and mapping the dynamic mechanical fingerprint into a biomechanical property map using an inverse dynamics solver, including:

[0117] The patient's lower limb length, interfemoral width and tibial plateau inclination are collected as lower limb biological parameters, and a parameterized knee joint mesh model is constructed based on the lower limb biological parameters. The parameterized knee joint mesh model includes the meniscus curvature radius and cruciate ligament attachment point coordinates that match the anatomical structure; the mechanical phase trajectory data is subjected to trajectory smoothing, and the patellar sliding velocity vector and acceleration change rate in the mechanical phase trajectory are extracted to generate a mechanical phase eigenvector; the principal component analysis is performed on the six-degree-of-freedom contact force distribution data in the joint cavity, and the six-degree-of-freedom contact force distribution in the joint cavity is extracted. The mechanical phase characteristic vector and the contact force characteristic vector are vector-superimposed on each other, and a dynamic mechanical fingerprint is generated through a weighted fusion algorithm. The weighted fusion algorithm dynamically adjusts the weight ratio of the mechanical phase characteristic vector and the contact force characteristic vector according to the biological parameters of the lower limb; the dynamic mechanical fingerprint is input into the inverse dynamics solver, and a mechanical inversion calculation is performed based on the parameterized knee joint mesh model to solve the cartilage contact stress distribution gradient and the ligament tension change rate to generate a biomechanical property map.

[0118] In this embodiment, the mechanical phase eigenvector is generated by performing trajectory smoothing on the mechanical phase trajectory data and extracting the patellar sliding velocity vector and acceleration change rate. These features describe the dynamic behavior of the patella during movement. The contact force eigenvector is generated by performing principal component analysis (PCA) on the six-degree-of-freedom contact force distribution data in the joint cavity and extracting the normal force fluctuation amplitude and the tangential force coupling coefficient. These features reflect the contact force distribution between the various parts inside the joint. The inverse dynamics solver is a mathematical tool that performs mechanical inversion calculations based on a parameterized knee joint mesh model and dynamic mechanical fingerprints to solve the cartilage contact stress distribution gradient and the ligament tension change rate to generate a detailed biomechanical property map.

[0119] In an embodiment of the present application, first, the patient's lower limb biological parameters (such as lower limb length, interfemoral width, and tibial plateau inclination) are collected, and a parameterized knee joint mesh model is constructed based on these parameters. The model contains detailed anatomical information such as the meniscus curvature radius and the coordinates of the cruciate ligament attachment points. Next, the mechanical phase trajectory data is smoothed, the patellar sliding velocity vector and the acceleration change rate are extracted, and a mechanical phase eigenvector is generated. At the same time, principal component analysis is performed on the six-degree-of-freedom contact force distribution data in the joint cavity, the normal force fluctuation amplitude and the tangential force coupling coefficient are extracted, and the contact force eigenvector is generated. Then, the mechanical phase eigenvector and the contact force eigenvector are vector-superimposed, and a dynamic mechanical fingerprint is generated using a weighted fusion algorithm, in which the weight ratio is dynamically adjusted according to the lower limb biological parameters. Finally, the dynamic mechanical fingerprint is input into an inverse dynamics solver, and a mechanical inversion calculation is performed based on the parameterized knee joint mesh model to solve the cartilage contact stress distribution gradient and the ligament tension change rate, and generate a biomechanical property map.

[0120] The following is a specific embodiment:

[0121] An athlete presented with recurring knee pain. The physician first measured the patient's lower limb length, interfemoral width, and tibial plateau inclination. He then collected knee imaging and mechanical data using a multi-angle dynamic X-ray machine and a six-degree-of-freedom contact force sensor. Based on these lower limb biometrics, the system constructed a parameterized knee mesh model, including the meniscus curvature radius and cruciate ligament attachment coordinates aligned with the patient's anatomy. Next, the system smoothed the mechanical phase trajectory data, extracted the patellar slip velocity vector and acceleration rate, and generated a mechanical phase eigenvector. The results showed a significant increase in patellar slip velocity during certain specific movements, suggesting potential slip abnormality or injury risk. Furthermore, the system performed principal component analysis on the six-degree-of-freedom contact force distribution data within the joint cavity, extracted the normal force fluctuation amplitude and tangential force coupling coefficient, and generated a contact force eigenvector. This analysis revealed abnormal fluctuations in the contact force distribution within the joint during walking, suggesting the possibility of early injury or dysfunction. The system then superimposed the mechanical phase eigenvectors with the contact force eigenvectors and generated a dynamic mechanical fingerprint using a weighted fusion algorithm. The weight ratio was dynamically adjusted based on the patient's lower limb biometric parameters. This process ensured the rational combination of different eigenvectors and improved the accuracy of data analysis. Finally, the system input the dynamic mechanical fingerprint into an inverse dynamics solver, which performed a mechanical inversion calculation based on a parameterized knee mesh model. This solved the cartilage contact stress distribution gradient and the rate of change of ligament tension, generating a detailed biomechanical property map. This not only helped doctors accurately diagnose patients' knee problems but also provided a scientific basis for developing effective treatment plans. This approach significantly improved the accuracy and personalization of diagnosis, helping to improve patients' recovery speed and quality of life. The entire process demonstrated how multi-source data fusion technology can be used to enhance the comprehensiveness and accuracy of knee injury identification.

[0122] Figure 2 The present invention provides a schematic diagram of the structure of an artificial intelligence-assisted diagnosis system for a knee joint, as shown in FIG. Figure 2 As shown, the system includes:

[0123] a receiving module 21 for receiving a multi-angle dynamic X-ray image sequence and six-degree-of-freedom contact force distribution data within the joint cavity, dynamically tracking the multi-angle dynamic X-ray image sequence and combining the six-degree-of-freedom contact force distribution data within the joint cavity, encoding the patellar slip trajectory to generate mechanical phase trajectory data;

[0124] A collection module 22 is configured to collect biological parameters of the patient's lower limbs to construct a personalized knee joint model, perform vector superposition of the mechanical phase trajectory data with the six-degree-of-freedom contact force distribution data within the joint cavity to generate a dynamic mechanical fingerprint, and map the dynamic mechanical fingerprint into a biomechanical property map using an inverse dynamics solver;

[0125] Construction module 23 is used to collect the frequency characteristics of the patient's surface electromyographic signal and the tibial rotation angle during exercise, construct an electromyographic activation pattern marked with the exercise phase, and perform feature fusion of the biomechanical property map with the electromyographic activation pattern to generate a comprehensive profile of joint dynamics;

[0126] An input module 24 is configured to input the joint dynamics comprehensive profile into a pre-built joint abnormal dynamics propagation model, calculate the wave propagation path of the meniscus stress concentration area and the spatiotemporal evolution trajectory of the cruciate ligament deformation energy field using the stress wave conduction equation, and identify the compensatory injury pattern by analyzing the interference effect between the spatiotemporal evolution trajectory and the wave propagation path;

[0127] The adjustment module 25 is used to generate a diagnostic map based on the compensatory injury pattern, combined with the biomechanical property map, the myoelectric activation pattern and the comprehensive profile of joint dynamics, match the biomechanical constraints in the preset motion feature library, and adaptively modify the diagnostic map to generate an individualized rehabilitation strategy.

[0128] Figure 2 The artificial intelligence-assisted diagnosis system for knee joints can perform Figure 1 The implementation principle and technical effects of the AI-assisted diagnosis method for the knee joint described in the illustrated embodiment will not be elaborated on here. The specific manner in which each module and unit performs operations in the AI-assisted diagnosis system for the knee joint in the above embodiment has been described in detail in the embodiments of the method and will not be elaborated on here.

[0129] In one possible design, Figure 2 The artificial intelligence-assisted diagnosis system for knee joints of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0130] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0131] The processing component 32 is as follows Figure 1 The embodiment provides an artificial intelligence-assisted diagnosis method for the knee joint.

[0132] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0133] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0134] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0135] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0136] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0137] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0138] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 An artificial intelligence-assisted diagnosis method for a knee joint according to the embodiment shown.

[0139] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0140] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0141] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0142] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An artificial intelligence-assisted diagnosis method for knee joints, characterized in that: include: receiving a multi-angle dynamic X-ray image sequence and six-degree-of-freedom contact force distribution data within the joint cavity, dynamically tracking the multi-angle dynamic X-ray image sequence and combining it with the six-degree-of-freedom contact force distribution data within the joint cavity, encoding the patellar sliding trajectory to generate mechanical phase trajectory data; Collecting biological parameters of the patient's lower limbs to construct a personalized knee joint model, performing vector superposition of the mechanical phase trajectory data and the six-degree-of-freedom contact force distribution data within the joint cavity to generate a dynamic mechanical fingerprint, and mapping the dynamic mechanical fingerprint into a biomechanical property map using an inverse dynamics solver; The frequency characteristics of the patient's surface electromyographic signals and the tibial rotation angle during exercise are collected to construct an electromyographic activation pattern marked with the exercise phase, and the biomechanical property map is fused with the electromyographic activation pattern to generate a comprehensive profile of joint dynamics; The joint dynamics comprehensive profile is input into a pre-built joint abnormal dynamics propagation model, and the stress wave conduction equation is used to calculate the wave propagation path of the meniscus stress concentration area and the spatiotemporal evolution trajectory of the cruciate ligament deformation energy field. The compensatory injury pattern is identified by analyzing the interference effect between the spatiotemporal evolution trajectory and the wave propagation path; According to the compensation injury pattern, the biomechanical property map, the myoelectric activation pattern and the comprehensive joint dynamics profile are combined to generate a diagnostic map, match the biomechanical constraints in the preset motion feature library, and adaptively modify the diagnostic map to generate an individualized rehabilitation strategy; The method includes inputting the comprehensive joint dynamics profile into a pre-built joint abnormal dynamics propagation model, calculating the wave propagation path of the meniscus stress concentration area and the spatiotemporal evolution trajectory of the cruciate ligament deformation energy field using the stress wave conduction equation, and identifying the compensatory injury pattern by analyzing the interference effect between the spatiotemporal evolution trajectory and the wave propagation path, including: Performing hierarchical tensor decomposition on the comprehensive joint dynamics profile to obtain biomechanical tensor components and kinematic tensor components, constructing a meniscus stress propagation path based on a stress waveguide model for the biomechanical tensor components, and calculating dynamic correlation parameters between the fluctuation attenuation rate and the stress concentration factor in the meniscus stress propagation path; Establishing a ligament energy field evolution model of the viscoelastic deformation field for the kinematic tensor component, and extracting the spatiotemporal gradient and energy residence node coordinates of the ligament energy field evolution model; An interference field equation is established by dynamically coupling a stress waveguide model with a viscoelastic deformation field using the dynamic correlation parameters, the spatiotemporal gradient, and the energy retention node coordinates; a wavefield interference intensity distribution between the meniscus stress propagation path and the energy retention node coordinates is calculated based on the interference field equation; and a compensatory damage pattern vector is generated based on the three-dimensional coordinates of an abnormal resonance region in the wavefield interference intensity distribution and an energy overflow threshold; Performing modal matching on the compensatory injury pattern vector and the standard kinematic chain parameters in the motion feature library, and performing frequency domain weighted adjustment on the compensatory injury pattern vector based on the matching result to generate a compensatory injury pattern; The method collects biological parameters of the patient's lower limbs to construct a personalized knee joint model, performs vector superposition of the mechanical phase trajectory data and the six-degree-of-freedom contact force distribution data in the joint cavity to generate a dynamic mechanical fingerprint, and maps the dynamic mechanical fingerprint into a biomechanical property map using an inverse dynamics solver, including: Collecting the patient's lower limb length, interfemoral width, and tibial plateau inclination as lower limb biological parameters, and constructing a parametric knee joint mesh model based on the lower limb biological parameters, wherein the parametric knee joint mesh model includes the meniscus curvature radius and cruciate ligament attachment point coordinates that match the anatomical structure; Performing trajectory smoothing processing on the mechanical phase trajectory data, extracting the patellar slip velocity vector and acceleration change rate in the mechanical phase trajectory, and generating a mechanical phase feature vector; performing principal component analysis on the six-degree-of-freedom contact force distribution data within the joint cavity, extracting the normal force fluctuation amplitude and the tangential force coupling coefficient in the six-degree-of-freedom contact force distribution within the joint cavity, and generating a contact force eigenvector; Performing vector superposition of the mechanical phase eigenvector and the contact force eigenvector, and generating a dynamic mechanical fingerprint through a weighted fusion algorithm, wherein the weighted fusion algorithm dynamically adjusts the weight ratio of the mechanical phase eigenvector and the contact force eigenvector according to the biological parameters of the lower limb; The dynamic mechanical fingerprint is input into an inverse dynamics solver, and a mechanical inversion calculation is performed based on the parameterized knee joint mesh model to solve the cartilage contact stress distribution gradient and ligament tension change rate, thereby generating a biomechanical property map.

2. The method according to claim 1, characterized in that The method includes establishing an interference field equation through the dynamic correlation parameters, the spatiotemporal gradient, and the energy residence node coordinates, and dynamically coupling a stress waveguide model with a viscoelastic deformation field; calculating a wavefield interference intensity distribution between the meniscus stress propagation path and the energy residence node coordinates according to the interference field equation; and generating a compensatory damage pattern vector according to the three-dimensional coordinates of an abnormal resonance region in the wavefield interference intensity distribution and an energy overflow threshold. The method includes: Performing wave equation parameterized reconstruction on the meniscus stress propagation path, extracting the wave phase delay parameter and amplitude attenuation rate in the meniscus stress propagation path, performing energy gradient field decomposition on the energy residence node coordinates, and generating an energy density gradient vector and a residence time coefficient; A dynamic coupling kernel function is constructed based on the wave phase delay parameter and the energy density gradient vector, and a spatial grid map of the wavefield interference intensity distribution is generated by combining the dynamic coupling kernel function with a spatial convolution operation. Grid cells having energy density peaks higher than an overflow threshold are marked in the spatial grid map as abnormal resonance regions. Using the amplitude decay rate and the residence time coefficient, the product of the energy residence time coefficient and the fluctuation amplitude decay rate in the abnormal resonance area is calculated as the compensation damage intensity factor, and the angle between the three-dimensional coordinates of the abnormal resonance area and the tangent direction of the fluctuation propagation path is extracted as the damage space distribution parameter. The compensation damage intensity factor and the damage space distribution parameter are tensor-recombined to generate a compensation damage pattern vector.

3. The method according to claim 2, characterized in that The method comprises: constructing a dynamic coupling kernel function based on the wave phase delay parameter and the energy density gradient vector, generating a spatial grid map of the wavefield interference intensity distribution by combining the dynamic coupling kernel function with a spatial convolution operation, and marking grid cells with energy density peaks higher than an overflow threshold in the spatial grid map as abnormal resonance regions. Performing time series normalization processing on the fluctuation phase delay parameter to generate a phase delay weight coefficient, and performing orthogonal basis decomposition on the energy density gradient vector to obtain a normal energy gradient component and a tangential energy diffusivity; A dual-channel coupling kernel function is constructed based on the phase delay weight coefficient and the normal energy gradient component, the tangential energy diffusivity is used as a dynamic attenuation factor of the dual-channel coupling kernel function, a three-dimensional convolution kernel is generated according to the dynamic attenuation factor, a dynamic window parameter of the spatial convolution operation is defined according to the anatomical structure of the knee joint, and a sliding integral calculation is performed in the biomechanical spatial domain using the dynamic window parameter in combination with the three-dimensional convolution kernel to generate an interference intensity distribution matrix; The interference intensity distribution matrix is ​​subjected to multi-scale Gaussian filtering, the spatial topological structure of the energy density peak in the filtered matrix is ​​detected, and the topological units in the spatial topological structure that are continuously adjacent and have peak values ​​higher than an overflow threshold are clustered as abnormal resonance regions.

4. The method according to claim 3, characterized in that The method comprises: constructing a dual-channel coupling kernel function based on the phase delay weight coefficient and the normal energy gradient component, using the tangential energy diffusivity as a dynamic attenuation factor of the dual-channel coupling kernel function, generating a three-dimensional convolution kernel according to the dynamic attenuation factor, defining a dynamic window parameter of a spatial convolution operation according to the anatomical structure of the knee joint, and performing a sliding integral calculation in a biomechanical spatial domain using the dynamic window parameter in combination with the three-dimensional convolution kernel to generate an interference intensity distribution matrix, including: Performing a frequency domain transform on the phase delay weight coefficient to generate a phase delay spectrum, extracting the main frequency band energy ratio in the phase delay spectrum as the time modulation factor of the dual-channel coupling kernel function, performing a spatial Fourier transform on the normal energy gradient component to generate a normal energy wavenumber spectrum, and extracting the high-frequency attenuation slope in the normal energy wavenumber spectrum as the spatial modulation factor of the dual-channel coupling kernel function; Constructing a main kernel structure of a dual-channel coupling kernel function based on the temporal modulation factor and the spatial modulation factor, introducing the logarithmic decay rate of the tangential energy diffusivity as a dynamic attenuation factor of the dual-channel coupling kernel function, and generating a three-dimensional convolution kernel; The initial value of the sliding radius is dynamically adjusted according to the meniscus curvature radius, and the direction vector of the convolution step is defined in combination with the anatomical characteristics of the ligament direction to generate dynamic window parameters that match the biomechanical characteristics of the knee joint. Performing a sliding integral calculation in the biomechanical spatial domain using the three-dimensional convolution kernel and the dynamic window parameters, adjusting the attenuation factor weight of the three-dimensional convolution kernel according to the local energy density gradient during each sliding process, and generating an initial mapping of the interference intensity distribution matrix; The initial mapping is subjected to local energy equalization processing, and the second-order derivative characteristics of the energy density gradient in the equalized mapping are extracted. The sliding radius and step size parameters of the three-dimensional convolution kernel are adjusted according to the second-order derivative characteristics to generate an interference intensity distribution matrix.

5. The method according to claim 1, wherein The method collects the frequency characteristics of the patient's surface electromyographic signals and the tibial rotation angle during exercise, constructs an electromyographic activation pattern marked with the exercise phase, and performs feature fusion on the biomechanical property map and the electromyographic activation pattern to generate a comprehensive profile of joint dynamics, including: Perform multi-scale time-frequency decomposition on the surface electromyographic signals collected from patients during exercise, extract the energy concentration coefficient and phase synchronization index of specific frequency bands in the time-frequency domain, and generate electromyographic time-frequency feature vectors; collecting time series data of tibial rotation angle, calculating the rotational dynamic stability coefficient of the time series data by the angle change rate, and constructing the electromyographic activation pattern marked by the movement stage in combination with the electromyographic time-frequency feature vector; Performing spatial domain feature extraction on the biomechanical property atlas to obtain the cartilage contact stress distribution gradient and ligament tension change rate in the atlas, and generating a biomechanical feature vector; Performing multimodal feature alignment on the electromyographic activation pattern and the biomechanical eigenvector, aligning the time series of the movement phases using a dynamic time warping algorithm, and generating a time-synchronized multimodal feature matrix; A comprehensive joint dynamics profile is constructed based on the multimodal characteristic matrix, and the energy concentration coefficient in the myoelectric activation pattern is coupled with the stress distribution gradient in the biomechanical characteristic vector through a nonlinear mapping function to generate a comprehensive dynamics profile vector.

6. An artificial intelligence-assisted diagnosis system for knee joints, characterized in that: include: a receiving module, configured to receive a multi-angle dynamic X-ray image sequence and six-degree-of-freedom contact force distribution data within the joint cavity, dynamically track the multi-angle dynamic X-ray image sequence and, in combination with the six-degree-of-freedom contact force distribution data within the joint cavity, encode the patellar sliding trajectory to generate mechanical phase trajectory data; a collection module for collecting biological parameters of the patient's lower limbs to construct a personalized knee joint model, performing vector superposition of the mechanical phase trajectory data with the six-degree-of-freedom contact force distribution data within the joint cavity to generate a dynamic mechanical fingerprint, and mapping the dynamic mechanical fingerprint into a biomechanical property map through an inverse dynamics solver; A construction module is used to collect the frequency characteristics of the patient's surface electromyographic signals and the tibial rotation angle during exercise, construct an electromyographic activation pattern marked with the exercise phase, and perform feature fusion of the biomechanical property map with the electromyographic activation pattern to generate a comprehensive profile of joint dynamics; An input module is configured to input the joint dynamics comprehensive profile into a pre-built joint abnormal dynamics propagation model, calculate the wave propagation path of the meniscus stress concentration area and the spatiotemporal evolution trajectory of the cruciate ligament deformation energy field using the stress wave conduction equation, and identify the compensatory injury pattern by analyzing the interference effect between the spatiotemporal evolution trajectory and the wave propagation path; An adjustment module is configured to generate a diagnostic map based on the compensation injury pattern, the biomechanical property map, the myoelectric activation pattern, and the comprehensive joint dynamics profile, match the biomechanical constraints in a preset motion feature library, and adaptively modify the diagnostic map to generate an individualized rehabilitation strategy; The method includes inputting the comprehensive joint dynamics profile into a pre-built joint abnormal dynamics propagation model, calculating the wave propagation path of the meniscus stress concentration area and the spatiotemporal evolution trajectory of the cruciate ligament deformation energy field using the stress wave conduction equation, and identifying the compensatory injury pattern by analyzing the interference effect between the spatiotemporal evolution trajectory and the wave propagation path, including: Performing hierarchical tensor decomposition on the comprehensive joint dynamics profile to obtain biomechanical tensor components and kinematic tensor components, constructing a meniscus stress propagation path based on a stress waveguide model for the biomechanical tensor components, and calculating dynamic correlation parameters between the fluctuation attenuation rate and the stress concentration factor in the meniscus stress propagation path; Establishing a ligament energy field evolution model of the viscoelastic deformation field for the kinematic tensor component, and extracting the spatiotemporal gradient and energy residence node coordinates of the ligament energy field evolution model; An interference field equation is established by dynamically coupling a stress waveguide model with a viscoelastic deformation field using the dynamic correlation parameters, the spatiotemporal gradient, and the energy retention node coordinates; a wavefield interference intensity distribution between the meniscus stress propagation path and the energy retention node coordinates is calculated based on the interference field equation; and a compensatory damage pattern vector is generated based on the three-dimensional coordinates of an abnormal resonance region in the wavefield interference intensity distribution and an energy overflow threshold; Performing modal matching on the compensatory injury pattern vector and the standard kinematic chain parameters in the motion feature library, and performing frequency domain weighted adjustment on the compensatory injury pattern vector based on the matching result to generate a compensatory injury pattern; The method collects biological parameters of the patient's lower limbs to construct a personalized knee joint model, performs vector superposition of the mechanical phase trajectory data and the six-degree-of-freedom contact force distribution data in the joint cavity to generate a dynamic mechanical fingerprint, and maps the dynamic mechanical fingerprint into a biomechanical property map using an inverse dynamics solver, including: Collecting the patient's lower limb length, interfemoral width, and tibial plateau inclination as lower limb biological parameters, and constructing a parametric knee joint mesh model based on the lower limb biological parameters, wherein the parametric knee joint mesh model includes the meniscus curvature radius and cruciate ligament attachment point coordinates that match the anatomical structure; Performing trajectory smoothing processing on the mechanical phase trajectory data, extracting the patellar slip velocity vector and acceleration change rate in the mechanical phase trajectory, and generating a mechanical phase feature vector; performing principal component analysis on the six-degree-of-freedom contact force distribution data within the joint cavity, extracting the normal force fluctuation amplitude and the tangential force coupling coefficient in the six-degree-of-freedom contact force distribution within the joint cavity, and generating a contact force eigenvector; Performing vector superposition of the mechanical phase eigenvector and the contact force eigenvector, and generating a dynamic mechanical fingerprint through a weighted fusion algorithm, wherein the weighted fusion algorithm dynamically adjusts the weight ratio of the mechanical phase eigenvector and the contact force eigenvector according to the biological parameters of the lower limb; The dynamic mechanical fingerprint is input into an inverse dynamics solver, and a mechanical inversion calculation is performed based on the parameterized knee joint mesh model to solve the cartilage contact stress distribution gradient and ligament tension change rate, thereby generating a biomechanical property map.

7. A computing device, characterized in that It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement an artificial intelligence-assisted diagnosis method for a knee joint as described in any one of claims 1 to 5.

8. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, an artificial intelligence-assisted diagnosis method for a knee joint as described in any one of claims 1 to 5 is implemented.

Images