Data analysis system for multi-joint exercise rehabilitation monitoring

Abstract

The invention relates to the technical field of data analysis, and discloses a data analysis system for multi-joint exercise rehabilitation monitoring. The system collects three-dimensional motion data, myoelectricity activity signals and spatial position data of multiple joints in real time through a multi-mode sensor array. The system comprises a data acquisition module, a preprocessing module, a space-time analysis module and a result output module. The preprocessing module is used for carrying out noise removal, normalization and feature extraction on the original data; the space-time analysis module constructs a monitoring model based on a graph space-time network, and analyzes space-time characteristics of multi-modal data through dynamic joint relation modeling by adopting an intelligent chip. And the result output module generates a visual report including exercise coordination evaluation, injury risk prediction and rehabilitation progress judgment, and feeds back the visual report to the rehabilitation equipment in real time. Besides, the system also has distributed storage and real-time feedback functions, the accuracy and efficiency of rehabilitation monitoring are improved, and the problems of single data processing and inaccurate spatial-temporal feature capture in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to a data analysis system for multi-joint motion rehabilitation monitoring. Background Art

[0002] In the field of rehabilitation medicine, multi-joint motion rehabilitation monitoring of patients is the key to evaluating treatment effects and formulating personalized rehabilitation plans. However, traditional monitoring methods often rely on manual observation and manual recording, which is not only inefficient but also easily affected by subjective factors. With the continuous development of sensor technology and data analysis technology, the use of multi-modal sensor arrays to collect multi-joint motion data in real time and analyze it in combination with advanced algorithms has become an important means to improve the accuracy and efficiency of rehabilitation monitoring. However, most existing data analysis systems can only process single-modal data, lack the ability to comprehensively analyze multi-modal data, and have difficulty in accurately capturing the spatiotemporal characteristics of joint motion, which limits their application in multi-joint motion rehabilitation monitoring. Summary of the invention

[0003] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to provide a data analysis system for multi-joint motion rehabilitation monitoring to solve the problems of single data processing and inaccurate capture of spatiotemporal features in the prior art.

[0004] In order to solve the above technical problems, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a data analysis system for multi-joint motion rehabilitation monitoring, the system comprising: The data acquisition module collects three-dimensional motion data, myoelectric activity signals and spatial position data in real time during multi-joint motion through a multi-modal sensor array; The data preprocessing module is used to remove noise, normalize and extract multimodal features from the data collected by the data acquisition module; The spatiotemporal analysis module builds a multi-joint motion rehabilitation monitoring model based on the graph spatiotemporal network, and uses smart chips to perform spatiotemporal feature analysis on the multimodal data preprocessed by the data preprocessing module through dynamic joint relationship modeling; The result output module generates a visual report containing movement coordination assessment, injury risk prediction and rehabilitation progress judgment based on the analysis results of the spatiotemporal analysis module, and feeds back to the rehabilitation equipment terminal in real time.

[0005] Preferably, in a possible implementation manner of the first aspect, the multimodal sensor array includes: Inertial measurement unit, used to obtain the three-dimensional acceleration, angular velocity and angular acceleration of the joint; A surface electromyography sensor, used to collect electromyography activity signals at a sampling rate of not less than 1000 Hz; Optical motion capture markers record the spatial position and motion trajectory of joints with sub-millimeter accuracy.

[0006] Preferably, in a possible implementation manner of the first aspect, the data preprocessing module includes: Data cleaning unit, using median filtering and wavelet threshold denoising algorithm to eliminate high-frequency noise; Normalization unit, mapping the data to the [0,1] interval through minimum-maximum normalization; The feature extraction unit combines principal component analysis with time-frequency domain analysis methods to extract joint motion amplitude, frequency spectrum entropy and root mean square value of electromyographic activity signal.

[0007] Preferably, in a possible implementation manner of the first aspect, the multi-joint movement rehabilitation monitoring model is constructed using a graph structure, and the graph structure construction process includes: Modeling N joints of the human body as a set of graph nodes , node features are multimodal data ,in represents a real matrix, d is the feature dimension, and T is the time step; Defining the adjacency matrix , edge weight By physical connection strength Relevance to exercise Jointly determine:

[0008] Where N is the number of joints, is an adjustable parameter, The correlation between joint motion trajectory and myoelectric activity signal is calculated through mutual information. and are the i-th and j-th nodes, and For the i and j Node features.

[0009] Preferably, in a possible implementation of the first aspect, the multi-joint motion rehabilitation monitoring model includes: Input layer, receiving preprocessed multimodal time series data; The multimodal data fusion layer uses a multi-head self-attention mechanism to fuse inertial data, electromyographic activity signals, and optical data; The spatiotemporal graph convolutional network layer consists of K stacked ST-GCN modules, each of which performs spatial graph convolution. and time-expanded convolution, the expansion factor varies with the network depth Incremental, where is the activation function, is an adjacency matrix with self-connection, is the degree matrix, For the Layer spatial graph convolution, For the Layer trainable weight matrix; Adaptive joint relationship modeling layer, which generates the latent joint association matrix through a differentiable graph learning algorithm; The multi-task output layer consists of two branches, which output the injury risk probability and rehabilitation progress score respectively.

[0010] Preferably, in a possible implementation of the first aspect, the structure of the spatiotemporal graph convolutional network layer includes: Dynamic spatial convolution module with learnable weight matrix , combined with the dynamic adjacency matrix Update node characteristics:

[0011] Where ReLU represents the activation function, BN is the batch normalization layer, is the current node feature, is the updated feature, is the input feature dimension, is the output feature dimension; Multi-scale temporal convolution module, using causal convolution to extract joint motion timing features ; Residual connection structure, the output of each ST-GCN module is , where Dropout is the random loss ratio, It is the input of the ST-GCN module.

[0012] Preferably, in a possible implementation manner of the first aspect, extracting features from the preprocessed multimodal data includes: Align inertial data, electromyographic activity signals, and optical data by time window; Generate joint features through multimodal data fusion layer; Extracting spatiotemporal features through spatiotemporal graph convolutional network layer , the spatiotemporal feature is the current motion feature.

[0013] Preferably, in a possible implementation manner of the first aspect, the content output to the result output module includes: Joint movement synchronization index ,in and are the angles of the i-th joint and the j-th joint at time t respectively; Injury risk probability , for The transpose of represents the weight vector of the damage risk prediction branch; Rehabilitation Progress Score ,in is the dynamic time warping distance, is the maximum allowed difference threshold, It is a standard recovery feature.

[0014] Preferably, in a possible implementation manner of the first aspect, the result output module includes: 3D motion trajectory simulation unit, reconstructing the 3D model of joint motion based on optical capture data ,in is the joint angle parameter; A damage risk prediction unit outputs a damage risk probability according to the result of the damage risk branch; The rehabilitation progress evaluation unit compares the DWT distance between the current motion feature and the standard rehabilitation feature and outputs a rehabilitation progress score.

[0015] Preferably, in a possible implementation manner of the first aspect, the system further includes: Distributed storage unit, using Hadoop architecture to store raw data, pre-processed data and analysis results; Real-time feedback unit, when joint movement deviation Greater than the preset deviation When the resistance parameter of the rehabilitation equipment is adjusted by the PID controller, the resistance parameter adjustment formula is:

[0016] in is the angle error, is the current joint motion angle, To set a recovery trajectory, is the proportional gain coefficient, is the integral gain coefficient, represents the time integral of the historical deviation, represents the rate of change of deviation, is the differential gain coefficient.

[0017] The beneficial effect of the present invention is that through the heterogeneous data fusion of multimodal sensor arrays and the deep feature mining of graph spatiotemporal networks, full-dimensional dynamic analysis and accurate feedback of multi-joint motion rehabilitation monitoring are realized. The system combines inertial sensing, electromyographic signals and optical capture technology to build a cross-modal data synchronous acquisition and collaborative analysis mechanism, breaking through the limitations of traditional single-modal monitoring and comprehensively capturing the biomechanical characteristics of joint motion and muscle synergistic activation patterns.

[0018] The spatiotemporal network model based on dynamic graph structure adaptively constructs the functional coupling relationship between joints through the dual weight optimization of physical connection and motion association, significantly improves the analytical ability of multi-joint collaborative features under complex motion modes, and effectively solves the problem of insufficient modeling of nonlinear spatiotemporal features by traditional linear models. The spatiotemporal graph convolutional network combines multi-scale causal convolution and residual learning mechanism to deeply extract the cross-joint spatial correlation features of motion trajectory while ensuring temporal causality, and realizes the early identification and dynamic tracking of abnormal motion patterns in rehabilitation movements. The application of intelligent chips and the coordinated design of distributed storage architecture and real-time feedback mechanism ensure the efficient processing and low-latency transmission of massive multi-source data, providing instant intervention and dynamic adjustment capabilities for rehabilitation training. The visual evaluation report output by the system integrates multi-dimensional indicators such as motion coordination, injury risk and rehabilitation progress, and combines three-dimensional motion trajectory reconstruction technology to intuitively present the biomechanical state of joint movement and assist clinical decision-making. Through adaptive graph learning and multi-task joint optimization, the system significantly enhances the dynamic adaptability to the individualized rehabilitation process, providing a reliable basis for the formulation and optimization of precise rehabilitation plans.

[0019] The present invention achieves breakthroughs in data fusion depth, spatiotemporal feature modeling accuracy and feedback real-time performance, effectively solving core problems in the prior art such as single data processing, inaccurate spatiotemporal feature capture and delayed feedback, and promoting rehabilitation monitoring from static assessment to dynamic intelligent regulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 A structural diagram of a data analysis system for multi-joint motion rehabilitation monitoring is provided for this application.

[0022] Explanation of the accompanying drawings: 1-data acquisition module, 2-data preprocessing module, 3-time-space analysis module, 4-result output module. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] Embodiment 1: Figure 1 As shown, the present invention provides a data analysis system for multi-joint motion rehabilitation monitoring, the system comprising: The data acquisition module 1 collects three-dimensional motion data, electromyographic activity signals and spatial position data in real time during multi-joint motion through a multi-modal sensor array.

[0025] Specifically, the multimodal sensor array adopts a heterogeneous sensor fusion architecture, including inertial measurement units, surface electromyography sensors, and optical motion capture markers.

[0026] The inertial measurement unit is used to obtain the three-dimensional acceleration, angular velocity and angular acceleration of the joint. In this embodiment, a 9-axis MEMS inertial sensor of model Xsens MTw Awinda is used, which integrates a three-axis accelerometer (range )、Three-axis gyroscope (range ) and a three-axis magnetometer. They are installed at the distal femur (anatomical landmark: 10cm above the lateral epicondyle of the femur), the proximal tibia (outside the tibial tuberosity) of the lower limb, the distal humerus (5cm proximal to the olecranon process) of the upper limb, and the radial styloid process. A sampling rate of 200Hz is used to meet the Nyquist theorem for capturing gait frequencies ≤15Hz, and data is transmitted in real time in quaternion format, and sensor data is fused through Kalman filtering.

[0027] Surface electromyography sensors are used to collect muscle electrical activity signals. The muscle electrical activity signal is collected at a sampling rate of 2000 Hz. In this embodiment, the Delsys Trigno system is used for electromyographic data collection, with an electrode spacing of 20 mm and a common mode rejection ratio of ≥100 dB. The attachment positions are the rectus femoris of the lower limb (1 / 3 of the line connecting the anterior superior iliac spine to the upper edge of the patella), the long head of the biceps femoris (the midpoint of the line connecting the ischial tuberosity and the fibular head), and the erector spinae of the trunk (2 cm away from the L3 spinous process). The data are collected at a sampling rate of 2000 Hz and a bandwidth of 20-450 Hz, and the RMS value is calculated in real time.

[0028] Optical motion capture markers record the spatial position and motion trajectory of joints with submillimeter accuracy. In this embodiment, an 8-camera infrared optical system of Qualisys Oqus 7+ model is used, and the diameter of the markers is 14mm. The markers are arranged on the anterior superior iliac spine, greater trochanter of the femur, lateral epicondyle of the femur, tibial tuberosity, lateral malleolus of the lower limbs, and the acromion, lateral epicondyle of the humerus, radial styloid process, and base of the third metacarpal bone of the upper limbs. Data collection is performed with a static positioning error of ≤0.2mm, a dynamic tracking error of ≤0.5mm, and a sampling rate of 120Hz.

[0029] At the same time, this embodiment adopts a multimodal data synchronization mechanism, including hardware synchronization, software alignment, and coordinate system 1. Hardware synchronization: The PXIe-6674T timing module is used to generate a 10MHz reference clock, and all sensors are triggered synchronously through the BNC interface; software alignment: The data stream uses the IEEE 1588 precision time protocol to ensure that the time deviation of multi-source data is less than 5ms; coordinate system 1: Establish a global coordinate system (the origin is the projection point of the first metatarsal head, and the Z axis is vertical to the ground and upward), and realize the rigid conversion between the IMU local coordinate system and the optical system global coordinate system through the calibration plate.

[0030] For the collected kinematic data, inertial data compensation is performed and the gravity component is eliminated by using the quaternion differential method. The formula is: ,in is the original output triaxial acceleration, is the rotation matrix, which represents the rotation transformation of the current posture of the IMU relative to the global coordinate system. is the gravitational acceleration vector, The three-axis acceleration is used to eliminate the influence of gravity. Butterworth 4th-order low-pass filtering is used for optical data filtering. For the electromyographic activity signal, adaptive filtering based on IMU acceleration data is used to eliminate motion artifacts, and the double threshold method is used for muscle activation detection. At the same time, the human skeleton model recommended by ISB is used to calculate the Euler angle to calculate the joint angle, and the plantar pressure switch is used to mark the ground contact period and the swing period to divide the gait cycle.

[0031] The data preprocessing module 2 is used to remove noise, normalize and extract multimodal features from the original data.

[0032] Specifically, the data preprocessing module 2 includes a data cleaning unit, a normalization unit and a feature extraction unit.

[0033] The data cleaning unit first performs median filtering on the three-dimensional acceleration and angular velocity data collected by the inertial measurement unit, using a sliding median filter with a window length of 5 sampling points (corresponding to a 25ms time window) to eliminate the spike noise caused by the instantaneous interference of the sensor. For the electromyographic activity signal collected by the surface electromyographic sensor, a three-layer decomposition and soft threshold denoising algorithm based on the Daubechies 4 wavelet basis is used to set the threshold (in is the estimated value of the noise standard deviation, and N is the signal length), effectively suppressing high-frequency noise and retaining muscle activation characteristics. The optical motion capture data was processed by a Butterworth fourth-order low-pass filter with a cutoff frequency set to 6 Hz to eliminate high-frequency jitter noise while retaining key motion features in the gait cycle.

[0034] The normalization unit uses the minimum-maximum scaling method to independently map each channel of the multimodal sensor data to the [0,1] interval. For the three-axis acceleration data of the inertial measurement unit, the extreme value reference is updated every 10 minutes to dynamically adapt to sensor drift; the normalization range of the electromyographic activity signal is truncated according to the baseline noise level in the resting state (±3 times the standard deviation) to avoid interference from outliers. The normalization formula is: ,in To prevent division by zero errors, the normalized data is stored in IEEE 754 floating point format to maintain accuracy.

[0035] The feature extraction unit integrates principal component analysis (PCA) and time-frequency domain joint analysis methods. First, the multimodal data (including IMU acceleration, sEMG RMS value, and optical joint angle) are time-aligned and spatially registered, and the data stream is segmented using a sliding window (window length 200ms, overlap rate 50%). The time domain features extract the joint motion amplitude (RMS value of the acceleration signal) and the integrated electromyographic value (iEMG) of the electromyographic activity signal; the frequency domain features calculate the power spectrum entropy, and the signal power spectrum density is estimated by the Welch method, with a segment length of 256 points and Hanning window weighting. The principal component analysis uses the cumulative variance contribution rate ≥95% as the criterion to reduce the high-dimensional feature vector to 12 dimensions and eliminate data redundancy. After feature fusion, the joint feature matrix containing joint kinematics, dynamics, and muscle activation patterns is output.

[0036] The multimodal data synchronization is hardware triggered by the 10MHz reference clock of the PXIe-6674T timing module. The software layer uses the IEEE 1588 precision time protocol to align the data stream, and the time deviation is controlled within 5ms. In the process of coordinate system normalization, the global coordinate system of the optical capture system is used as the reference (origin: the projection point of the first metatarsal head, the Z axis is perpendicular to the ground), and the rigid transformation matrix of the IMU local coordinate system is calculated by singular value decomposition (SVD) to ensure the spatial consistency of motion parameters.

[0037] In this implementation, the median filter and wavelet denoising algorithms of the data cleaning unit are implemented through embedded C++ code, calling the Intel MKL library to accelerate matrix operations; the normalization and feature extraction modules are run in the Python environment, using the Scipy and Scikit-learn libraries for real-time processing. The processed feature data is encapsulated in JSON format and uploaded to the cloud analysis platform via the wireless transmission module.

[0038] The spatiotemporal analysis module 3 builds a multi-joint motion rehabilitation monitoring model based on the graph spatiotemporal network, and uses smart chips to perform spatiotemporal feature analysis on the pre-processed multimodal data through dynamic joint relationship modeling.

[0039] Specifically, this embodiment adopts the NVIDIA Jetson AGX Orin model chip to achieve efficient spatiotemporal modeling by dynamically allocating GPU (processing joint topology spatial convolution) and CPU (updating dynamic adjacency matrix and sensor scheduling) tasks.

[0040] The multi-joint exercise rehabilitation monitoring model is constructed using a graph structure. The graph structure construction process includes: modeling N joints of the human body as a set of graph nodes , node features are multimodal data , where d is the feature dimension and T is the time step. Define the adjacency matrix , edge weight By physical connection strength Relevance to exercise Jointly determine:

[0041] in is an adjustable parameter, The correlation between joint motion trajectory and electromyographic activity signal is calculated through mutual information.

[0042] The multi-joint motion rehabilitation monitoring model includes an input layer, a multimodal data fusion layer, a spatiotemporal graph convolutional network layer, an adaptive joint relationship modeling layer, and a multi-task output layer.

[0043] The input layer receives preprocessed multimodal time series data , where N is the number of joints, d is the feature dimension, and T is the time window length. The preprocessed multimodal data input process includes: , electromyographic activity signals and optical data Align by time window; generate joint features through cross-modal fusion layer ; Extract spatiotemporal features through spatiotemporal graph convolutional network .

[0044] The multimodal data fusion layer uses a multi-head self-attention mechanism to fuse inertial data Q, electromyographic activity signal K, and optical data V:

[0045] in, is the dimension scaling factor of the key vector.

[0046] In this embodiment, the multimodal data fusion layer is implemented using the MultiHeadCrossAttention class, and its core function is to fuse heterogeneous data from inertial measurement units (IMUs), surface electromyography (sEMG), and optical capture systems. By defining a linear transformation layer (nn.Linear) of query (Q), key (K), and value (V) vectors, each modality data is mapped to a unified feature space. The scaled dot product attention calculation is performed using the built-in nn.MultiheadAttention component of PyTorch, where the query vector is generated by inertial data, and the key and value vectors correspond to electromyographic activity signals and optical data, respectively. The multi-head mechanism achieves feature capture in multiple semantic spaces by running four independent attention heads in parallel (n_heads=4), and the final output dimension remains a 64-dimensional fused feature vector.

[0047] The spatiotemporal graph convolutional network layer consists of K stacked ST-GCN modules, each of which performs spatial graph convolution. and time-expanded convolution, the expansion factor varies with the network depth Incremental, where is the activation function, is an adjacency matrix with self-connection, is the degree matrix, For the Layer spatial graph convolution, For the Layer trainable weight matrix.

[0048] The structure of the spatiotemporal graph convolutional network layer includes a dynamic spatial convolution module, a multi-scale temporal convolution module, and a residual connection structure.

[0049] Dynamic spatial convolution module with learnable weight matrix , combined with the dynamic adjacency matrix Update node characteristics: , where BN is the batch normalization layer.

[0050] Multi-scale temporal convolution module, using dilation factors Causal convolution to extract joint motion timing features: , k is the temporal convolution layer number, and d is the dilation factor of the causal convolution.

[0051] Residual connection structure, the output of each ST-GCN module is , where BN is batch normalization, Dropout is the random loss ratio, It is the input of the ST-GCN module.

[0052] In this embodiment, the spatiotemporal graph convolution is implemented by the STGCNBlock class, which includes dual feature processing in the spatial domain and the temporal domain. The spatial graph convolution uses the GCNConv graph convolution layer to propagate node features based on predefined physical connection edge indexes (such as the lower limb hip-knee-ankle chain connection). Its adjacency matrix generates anatomically constrained binary connection relationships through the _get_physical_edges method. The temporal convolution part uses nn.Conv1d to construct a causal dilated convolution, and the dilation factor increases exponentially with the number of network layers (the kth layer is ), padding = dilation is used to ensure temporal causality. Spatial convolution is performed frame by frame by looping over the time steps (for t in range(T)), followed by dilated convolution along the time dimension, and finally output features through batch normalization (nn.BatchNorm1d) and ReLU activation function. The residual connection structure (x_temporal + x) retains the original input information to alleviate the gradient vanishing problem.

[0053] Adaptive joint relationship modeling layer, generating potential joint association matrix through differentiable graph learning algorithm In this embodiment, the AdaptiveGraphLearner class implements the learning function of dynamic joint association weights. This module receives the node features output by the spatiotemporal graph convolution and calculates the potential association strength between joints through a fully connected network (nn.Sequential). Specifically, for each pair of joint nodes , concatenate their feature vectors and input them into two layers of MLP (32-dimensional hidden layer), and normalize them to edge weights of [0,1] by Sigmoid function. Double loops are used to traverse all joint pairs to generate an N×N weight matrix, which is linearly mixed with the predefined physical connection matrix in a ratio of 7:3 to form the final adjacency matrix.

[0054] The multi-task output layer contains two branches, which output the damage risk probability respectively. and rehabilitation progress score. Injury risk probability , is the weight vector of the injury risk prediction branch; rehabilitation progress score ,in is the dynamic time warping distance, is the maximum allowed difference threshold, It is a standard recovery feature.

[0055] The result output module 4 generates a visual report including movement coordination assessment, injury risk prediction and rehabilitation progress judgment, and feeds back to the rehabilitation equipment terminal in real time.

[0056] Specifically, the content input by the result output module 4 includes the joint movement synchronization index ,in and are the angles of the i-th joint and the j-th joint at time t, respectively, T is the length of the time window; the injury risk probability , is the weight vector of the injury risk prediction branch; rehabilitation progress score ,in is the dynamic time warping distance, is the maximum allowed difference threshold, It is a standard recovery feature.

[0057] The result output module 4 includes a three-dimensional motion trajectory simulation unit, an injury risk prediction unit and a rehabilitation progress assessment unit.

[0058] The 3D motion trajectory simulation unit reconstructs the 3D model of human motion using the inverse kinematics algorithm based on the coordinate data of the marker points of the optical capture system and the joint angle parameters of the inertial measurement unit. A personalized bone model is built through the OpenSim biomechanics platform, and the spatial trajectories of key marker points such as the lateral epicondyle of the femur and the radial styloid process are input into the model. The weighted least squares method is used to solve the joint rotation angle to generate a 3D motion simulation animation. The simulation results are rendered by the Unity3D engine, which supports perspective rotation, motion speed adjustment (0.1-2 times the speed) and key frame marking functions, and abnormal motion clips (in this embodiment, synchronization scoring) are ) is automatically highlighted in red. You can select a specific joint (such as the affected knee joint) through the touch interface to view the six-degree-of-freedom motion parameter curve, including the flexion / extension angle, valgus / valgus angle, and rotation angle over time.

[0059] The damage risk prediction unit integrates the risk probability output by the spatiotemporal analysis module 3 Combined with real-time biomechanical parameters, dynamic risk heat maps are generated. When the system automatically triggers the three-level warning mechanism: Level 1 warning Display a yellow warning icon on the visual interface; Level 2 warning Send speed reduction instructions to rehabilitation equipment via CAN bus; three-level warning The training equipment was immediately paused and an audible and visual alarm was sounded. The risk prediction results were correlated with biomechanical parameters (such as the peak contact force of the knee joint and the delay time of the erector spinae muscle activation). The Gaussian mixture model was used to perform cluster analysis on high-risk movement patterns, and the risk areas (such as the stress concentration area of ​​the anterior cruciate ligament of the femur) were marked in the three-dimensional simulation model.

[0060] The rehabilitation progress assessment unit uses the dynamic time warping (DTW) algorithm to calculate the patient's motion characteristics Standard recovery curve Similarity, output standardized score The standard curve library contains 200 groups of healthy subjects' data, which are stored by age (±5 years), BMI (±3), and gender. The most suitable reference group is automatically matched when the score is calculated. The evaluation results are visualized in the form of radar charts, and comparative analysis is performed from six dimensions, including movement coordination, muscle balance, and joint range of motion. When the improvement is less than 5%, the system automatically generates suggestions for adjusting the training program, including increasing the difficulty of the virtual reality task or adjusting the exoskeleton assistance parameters.

[0061] The visual report generation engine uses a template design to automatically compile the analysis results into a structured medical report. The report body contains: Movement coordination assessment: Display the synchronicity matrix of each joint pair. Abnormal values ​​(such as hip-knee synchrony < 0.5) are annotated with clinical explanations (such as "suggesting abnormal flexion-extension coordination pattern").

[0062] Injury risk distribution: Combine the heat map with the probability curve to mark the high-risk periods and their corresponding movement phases (such as the heel impact peak at 63% in the gait cycle); Rehabilitation Progress Tracking: Display The curve changes over time and the results of clinical scales such as Fugl-Meyer score are superimposed for double verification. The report output supports PDF / A medical archive format and HL7 FHIR standard interface, and is stored in the medical cloud platform after 256-bit AES encryption to ensure data privacy in compliance with GDPR regulations.

[0063] Embodiment 2: The present invention provides a data analysis system for multi-joint exercise rehabilitation monitoring, the system also includes a distributed storage unit and a real-time feedback unit Specifically, the distributed storage unit adopts a hybrid architecture based on the Hadoop ecosystem to achieve efficient storage and parallel processing of multimodal data. The raw sensor data (including IMU raw acceleration, sEMG time domain signal, and optical marker coordinates) are stored in the HDFS distributed file system in Apache Avro binary format. The data block size is set to 128MB to optimize MapReduce processing performance, and the replication factor is configured to 3 to ensure fault tolerance. The preprocessed feature data and spatiotemporal analysis results are stored in the HBase columnar database, and the row key is designed as "patient ID_timestamp_joint number", which supports millisecond-level time range queries. To meet the needs of clinical real-time analysis, the Alluxio memory acceleration layer is used to cache high-frequency access data for nearly 24 hours, and the cache replacement is managed by the LRU strategy. In terms of data security, the AES-256 encryption algorithm is used to encrypt static data, and the Kerberos protocol is combined to achieve fine-grained access control, which complies with HIPAA and GDPR medical data privacy regulations. The data lifecycle management module automatically migrates raw data over 180 days to Glacier low-cost storage, while retaining the permanent accessibility of feature data and evaluation results.

[0064] The real-time feedback unit realizes the adaptive resistance adjustment of the rehabilitation equipment through the fuzzy PID controller. Its core algorithm is: ,in is the angle error, is the current joint motion angle, To set a recovery trajectory, is the proportional gain coefficient, which determines the intensity of the system's response to the current deviation. is the integral gain coefficient, which determines the system's ability to eliminate steady-state errors. represents the time integral of the historical deviation, Indicates the deviation change rate, reflecting the joint movement trend. is the differential gain coefficient, which determines the system's ability to suppress sudden motion changes.

[0065] In this embodiment, the gain coefficient is dynamically adjusted according to the patient's motor ability: when the rehabilitation progress score When the conservative control strategy is adopted ( , , );when When , switch to the aggressive strategy ( , , ). The angle deviation data is smoothed by Kalman filtering, and the sampling interval is 5ms to ensure real-time control.

[0066] The resistance adjustment of the rehabilitation equipment is realized through the EtherCAT industrial bus, and the control instruction format follows the CiA 402 motion control specification. and lasts for 200ms), the system implements three-level braking strategy: Primary adjustment: Send 0x6041 status word command to linearly increase the device resistance torque to Calculated value; Intermediate intervention: If the deviation is not corrected within 500ms, the control mode switch 0x6040 is triggered to enable position-force hybrid control; Emergency brake: When When the device joints are locked, the 0x6042 fast stop instruction is used.

[0067] The system integration and communication architecture uses the OPC UA protocol to implement data interaction between modules and defines the following information model: Device node: contains real-time parameters such as joint angle, resistance torque, control mode, etc. Patient node: stores historical assessment reports, personalized control parameters and safety thresholds; Alarm node: records the timestamp, triggering reason and device response log of the braking event.

[0068] Real-time data streams are distributed through Apache Kafka message queues, and the topic partitioning strategy is allocated by patient ID hash to ensure that the sensor data, control instructions and feedback signals of the same patient maintain sequential consistency.

[0069] The safety monitoring mechanism includes a dual redundant design: the main controller uses the NI cRIO-9045 real-time processor to run the VxWorks system, and the backup controller is built on a Raspberry Pi 4. The two use heartbeat packets to achieve fault switching (switching delay <50ms). All control instructions must be digitally signed and verified (ECDSA algorithm) to prevent unauthorized parameter modifications. The device self-check module performs full system diagnosis every 30 minutes, and the detection items include sensor communication delay, storage unit CRC check, and actuator impedance test.

[0070] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A data analysis system for multi-joint exercise rehabilitation monitoring, characterized in that: The system comprises: The data acquisition module collects three-dimensional motion data, electromyographic activity signals and spatial position data in real time during multi-joint motion through a multi-modal sensor array; The data preprocessing module is used to remove noise, normalize and extract multimodal features from the data collected by the data acquisition module; The spatiotemporal analysis module builds a multi-joint motion rehabilitation monitoring model based on the graph spatiotemporal network, and uses smart chips to perform spatiotemporal feature analysis on the multimodal data preprocessed by the data preprocessing module through dynamic joint relationship modeling; The result output module generates a visual report containing movement coordination assessment, injury risk prediction and rehabilitation progress judgment based on the analysis results of the spatiotemporal analysis module, and feeds back to the rehabilitation equipment terminal in real time.

2. The data analysis system according to claim 1, characterized in that: The multimodal sensor array comprises: Inertial measurement unit, used to obtain the three-dimensional acceleration, angular velocity and angular acceleration of the joint; A surface electromyography sensor, used to collect electromyography activity signals at a sampling rate of not less than 1000 Hz; Optical motion capture markers record the spatial position and motion trajectory of joints with sub-millimeter accuracy.

3. The data analysis system according to claim 1, characterized in that: The data preprocessing module comprises: Data cleaning unit, using median filtering and wavelet threshold denoising algorithm to eliminate high-frequency noise; Normalization unit, mapping the data to the [0,1] interval through minimum-maximum normalization; The feature extraction unit combines principal component analysis with time-frequency domain analysis methods to extract joint motion amplitude, frequency spectrum entropy and root mean square value of electromyographic activity signal.

4. The data analysis system according to claim 1, characterized in that: The multi-joint motion rehabilitation monitoring model is constructed using a graph structure, and the graph structure construction process includes: Modeling N joints of the human body as a set of graph nodes , node features are multimodal data ,in represents a real matrix, d is the feature dimension, and T is the time step; Defining the adjacency matrix , edge weight By physical connection strength Relevance to exercise Jointly determine: Where N is the number of joints, is an adjustable parameter, The correlation between joint motion trajectory and myoelectric activity signal is calculated through mutual information. and are the i-th and j-th nodes, and For the i and j Node features.

5. The data analysis system according to claim 4, characterized in that: The multi-joint motion rehabilitation monitoring model comprises: Input layer, receiving preprocessed multimodal time series data; The multimodal data fusion layer uses a multi-head self-attention mechanism to fuse inertial data, electromyographic activity signals, and optical data; The spatiotemporal graph convolutional network layer consists of K stacked ST-GCN modules, each of which performs spatial graph convolution. and time-expanded convolution, the expansion factor varies with the network depth Incremental, where is the activation function, is an adjacency matrix with self-connection, is the degree matrix, For the Layer spatial graph convolution, For the Layer trainable weight matrix; Adaptive joint relationship modeling layer, which generates the latent joint association matrix through a differentiable graph learning algorithm; The multi-task output layer consists of two branches, which output the injury risk probability and rehabilitation progress score respectively.

6. The data analysis system according to claim 5, characterized in that: The structure of the spatiotemporal graph convolutional network layer includes: Dynamic spatial convolution module with learnable weight matrix , combined with the dynamic adjacency matrix Update node characteristics: Where ReLU represents the activation function, BN is the batch normalization layer, is the current node feature, is the updated feature, is the input feature dimension, is the output feature dimension; Multi-scale temporal convolution module, using causal convolution to extract joint motion timing features ; Residual connection structure, the output of each ST-GCN module is , where Dropout is the random loss ratio, It is the input of the ST-GCN module.

7. The data analysis system according to claim 6, characterized in that: Feature extraction of multimodal data after preprocessing includes: Align inertial data, electromyographic activity signals, and optical data by time window; Generate joint features through multimodal data fusion layer; Extracting spatiotemporal features through spatiotemporal graph convolutional network layer , the spatiotemporal feature is the current motion feature.

8. The data analysis system according to claim 7, characterized in that: The output to the result output module includes: Joint movement synchronization index ,in and are the angles of the i-th joint and the j-th joint at time t respectively; Injury risk probability , for The transpose of represents the weight vector of the damage risk prediction branch; Rehabilitation Progress Score ,in is the dynamic time warping distance, is the maximum allowed difference threshold, It is a standard recovery feature.

9. The data analysis system according to claim 8, characterized in that: The result output module includes: 3D motion trajectory simulation unit, reconstructing the 3D model of joint motion based on optical capture data ,in is the joint angle parameter; A damage risk prediction unit outputs a damage risk probability according to the result of the damage risk branch; The rehabilitation progress evaluation unit compares the DWT distance between the current motion feature and the standard rehabilitation feature and outputs a rehabilitation progress score.

10. The data analysis system according to claim 1, characterized in that: The system further comprises: Distributed storage unit, using Hadoop architecture to store raw data, pre-processed data and analysis results; Real-time feedback unit, when joint movement deviation Greater than the preset deviation When the resistance parameter of the rehabilitation equipment is adjusted by the PID controller, the resistance parameter adjustment formula is: in is the angle error, is the current joint motion angle, To set a recovery trajectory, is the proportional gain coefficient, is the integral gain coefficient, represents the time integral of the historical deviation, represents the rate of change of deviation, is the differential gain coefficient.

Images