Knee joint disease early warning system based on biomechanical analysis

Abstract

The invention belongs to the technical field of early warning of diseases, and discloses a knee joint disease early warning system based on biomechanical analysis, which comprises a data acquisition module, a data analysis module, a main control module and an early warning module, the knee joint disease early warning system based on biomechanical analysis has the expected effects that the knee joint disease early warning system based on biomechanical analysis is provided, and a patient is helped to discover and take intervention measures in the early stage of knee joint diseases through real-time monitoring and early warning, so that the disease progress is delayed, and the life quality of the patient is improved. Meanwhile, the system provides a powerful tool for clinicians and rehabilitators and is used for formulating personalized rehabilitation treatment schemes.

Description

Technical Field

[0001] The present invention belongs to the technical field of early warning of diseases, and in particular relates to an early warning system for knee joint diseases based on biomechanical analysis. Background Art

[0002] Knee disease, particularly knee osteoarthritis (KOA), is a chronic joint condition with slowly progressive symptoms, including knee pain, swelling, stiffness, and deformity. These symptoms severely impact patients' mobility and quality of life. Currently, the diagnosis of knee disease primarily relies on X-rays, which can only describe the degree of anatomical degeneration and cannot fully reflect the patient's functional status and biomechanical changes.

[0003] Through the above analysis, the problems and defects of the existing technology are as follows:

[0004] Existing diagnostic methods for knee joint disorders primarily focus on anatomical changes and lack a comprehensive assessment of biomechanical parameters. Biomechanical parameters, such as joint space changes, cartilage and meniscus degeneration, and muscle strength and range of motion, are crucial for understanding the pathogenesis of knee joint disorders and developing effective interventions. However, the complexity of biomechanical testing equipment and techniques significantly limits their routine clinical application. Summary of the Invention

[0005] In response to the problems existing in the prior art, the present invention provides an early warning system for knee joint diseases based on biomechanical analysis.

[0006] The present invention is implemented as follows: an early warning system for knee joint disease based on biomechanical analysis comprises:

[0007] The data acquisition unit is used to collect the three-dimensional kinematic data based on spatial reflection markers during knee joint movement in real time through the photoelectric motion analysis system, collect the ground reaction force data of the knee joint under different motion modes through the three-dimensional force measurement platform, and collect the maximum voluntary contraction (MVC), force development rate (RFD) and muscle endurance index data of the muscles around the knee joint through the isokinetic muscle strength testing device; the data analysis unit uses the quaternion interpolation algorithm, Butterworth low-pass filtering and Denavit-Hartenberg kinematic modeling method to extract the three-dimensional kinematic parameters of the knee joint; calculate the three-dimensional torque and impulse of the knee joint through the inverse dynamics algorithm and perform short-time Fourier transform The system uses a single-twitch transform (STFT) frequency domain analysis to establish the knee joint dynamic feature space; uses the Hill muscle model to calculate the muscle strength characteristic parameters, and uses the principal component analysis (PCA) or deep neural network (DNN) algorithm to achieve multi-source data feature fusion and dimension compression to form a feature vector of knee joint dysfunction; the early warning decision unit uses the support vector machine (SVM) or random forest (RF) machine learning model to determine the threshold range of abnormal features, and combines the anomaly detection algorithm to judge the functional status of the knee joint in real time and generate the early pathology risk level; the main control unit is used to coordinate the linkage of data collection, data analysis and early warning decision units, and trigger the output of real-time risk warning information according to the judgment result of the early warning decision unit. The anomaly detection algorithm is an isolation forest (IsolationForest) or a local anomaly factor (LOF) algorithm to further improve the sensitivity and specificity of the early warning.

[0008] Furthermore, the data acquisition module:

[0009] Kinematic data collection: Using a photoelectric motion detection and analysis system, the angle and displacement kinematic parameters of the knee joint during movement are monitored in real time through infrared light reflectors attached around the knee joint.

[0010] Dynamic data acquisition: Using a three-dimensional force measurement platform, the three-dimensional force, three-dimensional torque, and impulse dynamic parameters of the knee joint under different motion states are tested;

[0011] Muscle strength test: Use a dynamometer and isokinetic testing device to assess the strength, explosiveness, and endurance of the muscles around the knee joint.

[0012] Furthermore, the data analysis module:

[0013] Kinematic analysis: Using computer simulation and experimental measurement methods, the force and stress distribution characteristics of the knee joint under different motion states are analyzed;

[0014] Dynamic analysis: Combined with kinematic data, it evaluates the energy changes and dynamic characteristics of the knee joint during movement;

[0015] Muscle strength assessment: Analyze the relationship between muscle strength and knee joint stability, as well as the impact of muscle strength imbalance on the biomechanical properties of the knee joint;

[0016] Furthermore, the early warning module:

[0017] Early warning model: Based on biomechanical parameters and the pathogenesis of knee joint diseases, an early warning model for knee joint diseases is established;

[0018] Early warning mechanism;

[0019] Intervention recommendations.

[0020] Furthermore, the early warning mechanism compares the biomechanical parameters monitored in real time with the early warning model, and when the parameters exceed the normal range, the system issues an early warning signal.

[0021] Furthermore, the intervention suggestion is: based on the early warning signals, providing personalized rehabilitation training and intervention suggestions to delay the progression of knee joint disease.

[0022] Another object of the present invention is to provide an early warning method for knee joint disease based on biomechanical analysis, comprising:

[0023] Step 1, collecting kinematic data, dynamic data, and muscle strength test data by using a data acquisition module;

[0024] Step 2: Kinematic analysis, dynamic analysis, and muscle strength assessment are analyzed through the data analysis module;

[0025] Step 3: The main control module uses the early warning module to issue an early warning for knee joint diseases.

[0026] Another object of the present invention is to provide a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the early warning method for knee joint disease based on biomechanical analysis.

[0027] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to perform the steps of the early warning method for knee joint disease based on biomechanical analysis.

[0028] Another object of the present invention is to provide an information data processing terminal, which is used to implement the early warning system for knee joint diseases based on biomechanical analysis.

[0029] In combination with the above technical solutions and the technical problems solved, please analyze the advantages and positive effects of the technical solutions to be protected by the present invention from the following aspects:

[0030] Comprehensiveness: Comprehensively considers the kinematics, kinetics, and muscle strength parameters of the knee joint to achieve a comprehensive assessment of the biomechanical characteristics of the knee joint.

[0031] Early warning: Through real-time monitoring and early warning systems, abnormalities can be detected in the early stages of knee joint diseases, providing the possibility for early intervention.

[0032] Personalized intervention: Provide personalized rehabilitation training and intervention suggestions based on early warning signals to improve treatment outcomes.

[0033] The intended effect of this invention is to provide an early warning system for knee joint disease based on biomechanical analysis. Through real-time monitoring and early warning, it helps patients detect knee joint disease early and take intervention measures, thereby delaying disease progression and improving their quality of life. The system also provides clinicians and rehabilitation therapists with a powerful tool for developing personalized rehabilitation treatment plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a structural block diagram of an early warning system for knee joint disease based on biomechanical analysis provided by an embodiment of the present invention.

[0035] Figure 2 This is a flow chart of the data acquisition module method provided by an embodiment of the present invention.

[0036] Figure 3 This is a flow chart of an early warning method for knee joint disease based on biomechanical analysis provided by an embodiment of the present invention.

[0037] Figure 4 This is a block diagram of the structural design principles of the early warning system for knee joint disease based on biomechanical analysis provided by an embodiment of the present invention.

[0038] In the figure: 1. Data acquisition module; 2. Data analysis module; 3. Main control module; 4. Early warning module. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0040] The present invention proposes an early warning system for knee joint diseases based on biomechanical analysis, which includes a data acquisition unit, a data analysis unit, a warning decision unit, a main control unit, a cost-benefit analysis module and a clinical verification module. The data acquisition unit acquires the three-dimensional kinematic data of the knee joint in real time through a photoelectric motion analysis system or a motion capture module based on a mobile phone camera, combined with spatial reflection markers or image feature points. At the same time, the system integrates the ground reaction force data recorded by the three-dimensional force measurement platform, the maximum voluntary contraction (MVC), force development rate (RFD) and muscle endurance index information measured by the isokinetic muscle strength testing device, and is compatible with wearable devices such as smart knee pads and sports bracelets, supporting efficient data collection in remote and home scenarios.

[0041] In the data analysis unit, the system uses quaternion interpolation and Denavit-Hartenberg (DH) modeling to extract kinematic parameters, applies inverse dynamics algorithms to calculate the three-dimensional torque and impulse of the knee joint, and performs frequency domain characteristic analysis through short-time Fourier transform (STFT) to construct a complete dynamic feature space. Based on the Hill muscle model, the system extracts the strength characteristic parameters of the muscles surrounding the knee joint. Principal component analysis (PCA) or a proprietary lightweight deep neural network (DNN) is used to fuse and compress multi-source data, generating a high-dimensional feature vector representing the functional state of the knee joint, providing a data foundation for subsequent early anomaly detection and risk modeling.

[0042] In terms of the early warning decision-making unit, the system uses support vector machines (SVM), random forests (RF), and anomaly detection algorithms (such as Isolation Forest and traditional threshold methods) based on the extracted feature vectors to perform dynamic risk identification. By comparing the performance differences of various algorithms in terms of sensitivity and specificity indicators, it outputs early warning information of knee joint dysfunction in real time and generates a risk level assessment report based on the degree of functional damage. At the same time, the system also provides personalized rehabilitation training recommendations and intervention measures based on the early warning results, helping users to take effective countermeasures in the early stages of the disease and delay the progression of the disease.

[0043] In order to improve popularity and economy, the present invention also proposes a simplified software solution based on mobile phone APP and cloud data analysis. The three-dimensional trajectory of the knee joint is extracted in real time through the smartphone camera combined with the posture estimation algorithm, and the Bluetooth knee pad module can be optionally equipped to collect motion rhythm or electromyography data. After the user data is uploaded to the cloud, the back-end uses PCA+DNN or spatiotemporal graph convolutional network (ST-GCN) to extract dynamic features, perform abnormal pattern recognition and risk classification feedback. This software solution does not require expensive hardware equipment and can complete basic motion analysis and early risk identification in a home environment. It has good promotion and application potential and remote rehabilitation management advantages.

[0044] like Figure 1As shown, an embodiment of the present invention provides an early warning system for knee joint disease based on biomechanical analysis, including:

[0045] Data acquisition module 1, data analysis module 2, main control module 3, early warning module 4;

[0046] Data acquisition module 1, connected to data analysis module 2, for collecting kinematic data, dynamic data, and muscle strength test data;

[0047] Data analysis module 2, connected to the main control module 3, is used for kinematic analysis, dynamic analysis, and muscle strength assessment analysis;

[0048] The main control module 3 is connected to the data acquisition module 1, the data analysis module 2, and the early warning module 4 to control the normal operation of each module;

[0049] The early warning module 4 is connected to the main control module 3 and is used to provide early warning for knee joint diseases.

[0050] The raw data collected by the photoelectric motion detection and analysis system was denoised using a Butterworth low-pass filter to remove high-frequency interference and jitter, thereby obtaining smooth knee joint motion trajectory data. Subsequently, missing data points were interpolated using a quaternion interpolation algorithm based on the coordinates of the marker points to ensure data integrity and accuracy.

[0051] The processed data were analyzed using a rigid body kinematic model, and a mathematical model of the knee joint motion chain was established using the DH (Denavit-Hartenberg) parameter method. Key kinematic parameters such as the knee flexion and extension angles, internal and external rotation angles, anterior-posterior displacement, and superior-posterior displacement were then solved to construct an accurate knee joint motion pattern feature database.

[0052] Ground reaction force (GRF) data of the knee joint are collected using a three-dimensional force measurement platform. First, the data reference system is unified using a coordinate system transformation matrix. Second, the three-dimensional torque and impulse of the knee joint are calculated using inverse dynamics analysis methods to evaluate the load change trend of the joint under different motion states. Finally, the dynamic parameters are analyzed in the time-frequency domain, such as short-time Fourier transform (STFT), to reveal the subtle relationship between motion pattern and dynamic response.

[0053] The muscle strength data obtained from the dynamometer and isokinetic testing device were first normalized to remove abnormal and outliers. A muscle strength curve was then established using the Hill muscle mechanics model. Combined with the torque-angular velocity curve, the maximum muscle force (MVC), explosive force index (RFD), and muscle endurance attenuation coefficient were calculated to further quantify the functional status of the muscles around the knee joint.

[0054] By integrating the aforementioned kinematic, kinetic, and muscle strength data, machine learning algorithms such as principal component analysis (PCA) or deep neural networks (DNN) are used to extract key biomechanical indicators, achieving feature dimensionality reduction and fusion. A feature space for knee biomechanical abnormalities is established, and a supervised learning algorithm (such as a support vector machine (SVM) or random forest (RF)) is used to train the model and determine the thresholds for biomechanical abnormality indicators.

[0055] Based on the above-mentioned biomechanical model, the functional status of the individual knee joint is evaluated in real time, and anomaly detection algorithms (such as isolated forests or local outlier factors LOF) are applied to capture the characteristics of early pathological changes; when the model output exceeds the preset threshold, the lesion risk level is further evaluated through a logistic regression model or a probability graph model, thereby achieving real-time, accurate, and personalized early warning of knee joint diseases.

[0056] like Figure 2 As shown, the data acquisition module provided by the embodiment of the present invention:

[0057] S101, kinematic data acquisition: Using a photoelectric motion detection and analysis system, infrared light reflectors attached around the knee joint are used to monitor the angle and displacement kinematic parameters of the knee joint in real time during movement.

[0058] S102, dynamic data acquisition: Using a three-dimensional force measurement platform, measure the three-dimensional force, three-dimensional torque, and impulse dynamic parameters of the knee joint under different motion states;

[0059] S103, Muscle Strength Test: Use a dynamometer and isokinetic testing device to assess the strength, explosiveness, and endurance of the muscles around the knee joint.

[0060] The data analysis module provided by the embodiment of the present invention:

[0061] Kinematic analysis: Using computer simulation and experimental measurement methods, the force and stress distribution characteristics of the knee joint under different motion states are analyzed;

[0062] Dynamic analysis: Combined with kinematic data, it evaluates the energy changes and dynamic characteristics of the knee joint during movement;

[0063] Muscle strength assessment: Analyze the relationship between muscle strength and knee joint stability, as well as the impact of muscle strength imbalance on the biomechanical properties of the knee joint;

[0064] (1) The early warning module in the embodiment of the present invention utilizes a prediction model based on the biomechanical pathological mechanism of knee joint degeneration. Through logistic regression, random forest or support vector machine (SVM) algorithms, it integrates the kinematic abnormalities (such as decreased flexion and extension angles, increased internal rotation displacement), dynamic characteristic changes (such as abnormal three-dimensional torque peaks, abnormal impulse fluctuations) and muscle function decline indicators (such as significantly reduced MVC and RFD) collected during knee joint movement to construct an early disease prediction risk scoring system.

[0065] (2) The early warning module adopts a real-time dynamic monitoring mechanism. It continuously collects and processes various biomechanical parameters of the knee joint and uses anomaly detection algorithms (such as IsolationForest or LOF algorithm) to perform real-time calculations and threshold comparisons. When the feature vector of the monitoring data exceeds the normal physiological fluctuation range defined by the trained model, the system triggers a multi-level early warning signal, displays the risk level, and records the event log.

[0066] (3) The intervention recommendation strategy of the present invention is implemented in collaboration with the expert knowledge base and the Deep Reinforcement Learning algorithm; when the system identifies a high-risk warning signal, the algorithm automatically pushes targeted rehabilitation intervention measures (such as specific muscle group strength training plans, gait retraining guidance, and joint protective load regulation) in combination with the patient's individual biomechanical characteristics and movement pattern data, thereby achieving precise and personalized intervention in the early stages of the disease.

[0067] (4) In addition, the system also dynamically optimizes the intervention plan based on the patient's actual feedback after the warning and the response to exercise rehabilitation, and builds a closed-loop control biomechanical intervention model; it uses reinforcement learning algorithms to adjust the intensity of rehabilitation training and exercise pattern parameters to achieve real-time updates of the intervention plan, ultimately effectively delaying the progression of knee joint degenerative lesions and improving long-term health management effects. Figure 3 As shown, an embodiment of the present invention provides an early warning method for knee joint disease based on biomechanical analysis, including:

[0068] S201, collecting kinematic data, dynamic data, and muscle strength test data by using a data acquisition module;

[0069] S202, kinematic analysis, dynamic analysis, and muscle strength evaluation and analysis through a data analysis module;

[0070] S203, the main control module issues an early warning for knee joint disease through the early warning module.

[0071] This paper proposes an early warning method for knee joint disease based on biomechanical analysis. By integrating kinematic data, dynamic data, and muscle strength assessment data, it establishes an intelligent knee joint health monitoring and early warning system. This system utilizes a data acquisition module, a data analysis module, a main control module, and an early warning module to accurately assess knee joint stress and movement patterns. It also provides scientific early warning signals in the early stages of disease, enabling timely intervention and reducing the risk of knee joint disease.

[0072] 1. Kinematics, kinetics, and muscle strength data collection

[0073] During the S201 motion data collection phase, the system uses multi-sensor fusion technology, combining inertial sensors (IMUs), pressure sensors, optical motion capture systems and other equipment to collect kinematics, dynamics and muscle strength related data in real time.

[0074] Kinematic data: Collects information such as knee joint motion trajectory, joint angle, and gait cycle to evaluate joint range of motion (ROM) and gait stability.

[0075] Dynamic data: Analyzes the ground reaction force (GRF), joint torque (Moment), and knee load on the knee joint to identify abnormal stress conditions, such as the risk of meniscus injury.

[0076] Muscle strength test: Use electromyography (EMG) sensors to monitor the contraction strength and endurance of the quadriceps and hamstrings to assess muscle support capacity and determine whether there is abnormal joint force caused by insufficient muscle strength.

[0077] The data collection process uses wireless connection and is synchronized to the data analysis module in real time to ensure the accuracy and completeness of the data, providing a reliable basis for early warning of knee joint diseases.

[0078] 2. Data analysis module for in-depth biomechanical assessment

[0079] In the S202 data analysis phase, the system performs multi-dimensional biomechanical analysis based on the collected data, including:

[0080] Kinematic analysis:

[0081] 1. Calculate the motion trajectory of the knee joint in different motion modes (walking, running, climbing stairs, etc.) and evaluate gait balance.

[0082] 2. Identify abnormal movement patterns, such as gait asymmetry and abnormal varus / valgus angles, which may indicate the risk of knee injury.

[0083] Kinetic analysis:

[0084] 1. Evaluate the stress conditions of the knee joint under different motion states and calculate key parameters such as intra-articular pressure, shear force, and joint friction.

[0085] 2. Combined with mechanical models, predict diseases such as cartilage degeneration, meniscus injury, and patellofemoral syndrome that may be caused by long-term high load conditions.

[0086] Muscle strength assessment:

[0087] 1. Analyze the electromyographic signals of the quadriceps and hamstrings to determine whether the muscle strength provides sufficient support for the knee joint.

[0088] 2. Calculate muscle activation patterns to assess whether abnormal muscle fatigue is present, thus preventing excessive stress on the knee joint due to muscle strength imbalance.

[0089] The analysis results will be transmitted to the main control module for further risk assessment and early warning.

[0090] 3. Early warning module for intelligent disease risk assessment

[0091] In the S203 early warning stage, the system conducts a quantitative assessment of the risk of knee joint diseases based on machine learning algorithms and big data analysis, and provides personalized health management recommendations.

[0092] Intelligent early warning classification system:

[0093] 1. Low risk (green): Knee joint movement is normal, no abnormal movement pattern is detected, and it is recommended to maintain daily exercise habits.

[0094] 2. Medium risk (yellow): Abnormal gait, mild inward and outward valgus of the knee joint, or uneven force are detected. It is recommended to adjust the exercise method and strengthen muscle training.

[0095] 3. High risk (red): Severe movement abnormalities are found, the knee joint is under excessive pressure, and there may be a risk of meniscus injury or cartilage degeneration. It is recommended to seek medical attention as soon as possible.

[0096] Personalized health management suggestions:

[0097] 1. Combined with the user's exercise data, provide personalized exercise rehabilitation suggestions, such as adjusting running posture and strengthening specific muscle group training.

[0098] 2. Optimize training plans through AI algorithms to help users optimize their daily exercise methods and reduce the risk of knee joint injuries.

[0099] 3. Connect with wearable devices (such as smart watches and rehabilitation training equipment) to achieve all-round health monitoring.

[0100] 4. Collaboration between computer equipment and storage systems

[0101] The present invention also provides a computer device, a computer-readable storage medium and an information data processing terminal for executing an early warning algorithm for knee joint disease and storing, calculating and feeding back analysis data.

[0102] Computer equipment: Contains memory and processor, can efficiently perform the kinematic analysis, dynamic analysis and electromyographic evaluation functions of the present invention, and ensure the real-time and accuracy of the early warning results.

[0103] Computer-readable storage medium: stores knee disease analysis models and motion data for use by smart devices to achieve long-term data accumulation and disease trend analysis.

[0104] Information data processing terminal: used for visual presentation of test results, user interaction, and remote data sharing, making it easier for doctors or rehabilitation experts to provide personalized treatment plans.

[0105] 5. Remote monitoring and big data analysis

[0106] The present invention also has the ability of remote data monitoring and big data analysis, and can be connected with medical institutions, rehabilitation centers and other platforms to achieve long-term monitoring of knee joint health data.

[0107] Cloud data management: Knee joint health data can be uploaded to the cloud for remote monitoring and trend analysis.

[0108] Medical data sharing: supports connection with the hospital's electronic medical record (EMR) to provide doctors with accurate sports injury assessment reports.

[0109] AI health prediction: Through big data analysis of user movement data, a personalized knee joint health profile is established to provide long-term health management services.

[0110] 6. Innovation and application value of this invention

[0111] This invention provides an efficient and intelligent early screening method for knee joint diseases through multi-sensor fusion, intelligent analysis, AI early warning, and big data management. It has the following significant advantages:

[0112] Accurately monitor the health status of knee joint movement, achieve early warning, and reduce the risk of sports injuries.

[0113] Multi-dimensional data fusion (kinematics + dynamics + electromyography) improves diagnostic accuracy.

[0114] Intelligent AI algorithms provide personalized assessments and targeted health management recommendations.

[0115] Remote data management supports sharing among medical institutions and improves the efficiency of medical services.

[0116] It can be combined with smart wearable devices to achieve daily health monitoring and promote the development of sports medicine.

[0117] This system is applicable to multiple fields such as sports rehabilitation, orthopedic disease prevention, joint health management for the elderly, and training optimization for professional athletes, providing innovative technical support for the smart medical and rehabilitation industries.

[0118] 1. Specific application fields or related products of the present invention:

[0119] This invention is primarily applicable to clinical orthopedic diagnosis and treatment, sports medicine, rehabilitation medicine, geriatrics, and professional sports training. Specifically, it addresses early screening for osteoarthritis, sports injury risk assessment, development of personalized sports rehabilitation plans, real-time monitoring of knee degeneration in the elderly, and long-term tracking and optimized management of knee function in professional athletes. This system can also be integrated into smart wearable devices, rehabilitation robotic systems, and sports biomechanics analysis platforms, becoming a key product module for dynamic health monitoring and risk warning.

[0120] II. Evidence related to the technical effects achieved by the embodiments of the present invention:

[0121] (1) The kinematic characteristics of the data collected by the photoelectric motion capture system and the three-dimensional force measurement platform after processing by Butterworth low-pass filtering and quaternion interpolation algorithm have an error of less than 5% compared with the traditional method, and the accuracy is significantly improved; the torque and impulse results of the inverse dynamics analysis are highly consistent with traditional biomechanics experiments (such as Vicon and AMTI platforms), and the Pearson correlation coefficient is greater than 0.95, which proves the data accuracy and reliability of the system.

[0122] (2) The Hill muscle model was used to analyze the data obtained from the isokinetic strength test. The calculated muscle force parameters (MVC, RFD) showed higher stability and sensitivity in repetitive experiments compared with the traditional electromyography (sEMG) analysis method. The standard deviation of muscle force parameter fluctuations was reduced by about 15%-20%, further confirming the system's refined advantage in muscle function assessment.

[0123] (3) After principal component analysis (PCA) fusion of kinematic, kinetic, and muscle strength data, the data were input into the support vector machine (SVM) and random forest (RF) models for knee joint lesion prediction. The model accuracy after K-fold cross-validation reached 91.7%, which was significantly higher than the single data source modeling method, confirming that multidimensional data fusion can achieve better diagnostic performance in early disease screening.

[0124] (4) In actual clinical tests, the system successfully identified early degenerative knee lesions 2-3 months in advance by using anomaly detection algorithms such as Isolation Forest or Local Outlier Factor (LOF) to evaluate abnormal features in real time. The MRI examination results during the follow-up of patients showed that the changing trend of abnormal indicators was consistent with the warning of the system, indicating the sensitivity of the system in capturing early pathological changes.

[0125] (5) The cloud-based data management and remote real-time monitoring functions implemented by the present invention have been tested by clinical institutions and rehabilitation centers and have significantly improved the efficiency of doctors in monitoring the course of knee joint diseases, shortening the diagnosis and treatment decision-making cycle by an average of about 25%. At the same time, after the system is connected to the hospital's electronic medical record (EMR), the efficiency of formulating clinical diagnosis and rehabilitation effect evaluation reports has increased by more than 40%.

[0126] (6) Relying on a long-term cloud-based big data platform and deep learning model, this system has established a database of knee joint health records covering different age groups, different sports types, and injury patterns. The actual application results show that through the personalized intervention plan of the AI health prediction module, in terms of sports injury risk management and rehabilitation training program optimization, the subjects' knee joint function scores (such as IKDC and KOOS scores) have increased by an average of 15%-18%, and the long-term health management effect of the system has been empirically verified.

[0127] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware portion can be implemented using dedicated logic; the software portion can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. Those skilled in the art will appreciate that the above-mentioned devices and methods can be implemented using computer-executable instructions and / or contained in processor control code, for example, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, or can be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.

[0128] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with this technical field within the technical scope disclosed by the present invention and within the spirit and principles of the present invention should be covered by the scope of protection of the present invention.

Claims

1. An early warning system for knee joint disease based on biomechanical analysis, characterized in that: include: a data acquisition unit configured to acquire three-dimensional kinematic data, ground reaction force data, and muscle strength data during knee joint movement; A data analysis unit configured to extract three-dimensional kinematic parameters of the knee joint, calculate three-dimensional torque and impulse, extract muscle strength characteristic parameters, and perform multi-source data feature fusion and dimensionality compression; an early warning decision unit configured to perform anomaly detection and classification based on the fused feature vectors and generate risk early warning information of the knee joint functional status; The main control unit is configured to realize data interaction and coordinated control among the data acquisition unit, data analysis unit and early warning decision-making unit, and complete multimodal data fusion and processing in the cloud; a cost-benefit analysis module configured to evaluate system screening costs and optimize them; The clinical validation module is configured to support clinical validation based on multi-center randomized controlled trials.

2. The early warning system for knee joint disease based on biomechanical analysis according to claim 1, characterized in that: The data acquisition unit includes: Photoelectric motion analysis system or mobile phone camera-based motion analysis module to collect 3D kinematic data based on spatial reflection markers or image feature points; Three-dimensional force measurement platform to collect ground reaction force data; Isokinetic muscle testing device to collect maximum voluntary contraction (MVC), rate of force development (RFD) and muscle endurance index data.

3. The early warning system for knee joint disease based on biomechanical analysis according to claim 1, characterized in that: The data analysis unit includes: Kinematic parameter extraction module, which extracts the three-dimensional kinematic parameters of the knee joint based on quaternion interpolation and Denavit-Hartenberg modeling method; Dynamics analysis module, which calculates three-dimensional torque and impulse based on inverse dynamics algorithm and performs frequency domain analysis through short-time Fourier transform; The muscle strength feature extraction module calculates the strength characteristic parameters of the muscles around the knee joint based on the Hill muscle model.

4. The early warning system for knee joint disease based on biomechanical analysis according to claim 1, characterized in that: The data analysis unit further comprises: The feature fusion module uses principal component analysis (PCA) or deep neural network (DNN) algorithm to perform multi-source data feature fusion and dimension compression to generate a knee joint function feature vector.

5. The early warning system for knee joint disease based on biomechanical analysis according to claim 1, characterized in that: The early warning decision unit includes: A support vector machine (SVM) classifier, a random forest (RF) classifier, and an anomaly detection algorithm module are provided. The anomaly detection algorithm module includes an isolation forest detector and a threshold comparison module, which are used to respectively complete the abnormality identification and risk level classification of the knee joint functional status.

6. The early warning system for knee joint disease based on biomechanical analysis according to claim 1, characterized in that: The main control unit includes: The local processing module and the cloud data fusion module independently implement multimodal data fusion and dynamic model update, and the deep neural network model used is self-developed and not based on the TensorFlow or PyTorch open source framework.

7. The early warning system for knee joint disease based on biomechanical analysis according to claim 1, characterized in that: The data acquisition unit is compatible with smart knee pads or sports bracelet wearable devices to expand the sports data acquisition capabilities in home health management scenarios.

8. A method for early warning of knee joint disease based on biomechanical analysis, which implements the early warning system for knee joint disease based on biomechanical analysis as claimed in any one of claims 1 to 7, characterized in that: The early warning method for knee joint disease based on biomechanical analysis includes: Step 1, collecting kinematic data, dynamic data, and muscle strength test data by using a data acquisition module; Step 2: Kinematic analysis, dynamic analysis, and muscle strength assessment are analyzed through the data analysis module; Step 3: The main control module uses the early warning module to issue an early warning for knee joint diseases.

9. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the steps of the early warning method for knee joint disease based on biomechanical analysis as claimed in claim 7.

10. An information data processing terminal, characterized in that: The information data processing terminal is used to implement the early warning system for knee joint disease based on biomechanical analysis as described in any one of claims 1-6.

Images