Quantitative evaluation method and system for exercise rehabilitation effect

Abstract

The invention belongs to the technical field of exercise rehabilitation, and particularly discloses an exercise rehabilitation effect quantitative evaluation method which comprises the following steps: S1, a data acquisition unit acquires data of a patient in an exercise process through fusion of a multi-source sensor; s2, muscle electrical activity signals in the movement process of the patient are collected through bio-electricity signals, and the activation degree and the fatigue condition of muscles in rehabilitation movement are known; s3, acquiring muscle electrical activity signal data according to the acquired multi-source fusion data and the bio-electricity signals, and performing rehabilitation evaluation on the model; s4, according to the output of the evaluation model, a quantitative rehabilitation effect evaluation report is generated, and the report content comprises numerical values of various quantitative indexes, comparative analysis of rehabilitation targets and comprehensive scores of rehabilitation effects; an exercise rehabilitation effect quantitative evaluation system comprises a data acquisition system and a rehabilitation evaluation module. The method solves the problems that a traditional evaluation method is high in subjectivity and single in data, and has the advantages of being objective, comprehensive and accurate in quantification.

Description

Technical Field

[0001] The present invention belongs to the technical field of sports rehabilitation, and in particular relates to a method and system for quantitatively evaluating sports rehabilitation effects. Background Art

[0002] With the improvement of people's health awareness and the increase in rehabilitation needs caused by sports injuries and chronic diseases, sports rehabilitation is receiving more and more attention. Traditional evaluation of sports rehabilitation effects mainly relies on the subjective judgment of rehabilitation physicians, such as observing the completion of the rehabilitation patients' movements and asking the rehabilitation patients about their subjective feelings. This evaluation method has problems such as strong subjectivity, insufficient accuracy, and difficulty in precise quantification. For example, when evaluating the range of motion of the rehabilitation patients, different physicians may have different judgments, and it is impossible to accurately quantify the degree of improvement in joint mobility. In addition, some existing evaluation methods based on single sensor data cannot fully reflect the rehabilitation status of the rehabilitation patients due to limited data information. Therefore, it is of great practical significance to develop a method and system that can accurately, comprehensively, and quantitatively evaluate the effects of sports rehabilitation. Summary of the Invention

[0003] The purpose of this application is to provide a method and system for quantitative evaluation of exercise rehabilitation effects, which has the advantages of achieving comprehensive and objective quantitative evaluation of rehabilitation effects through multi-source sensor fusion and bioelectric signal acquisition, thereby improving the accuracy and quantification of the evaluation.

[0004] This application provides a method for quantitatively evaluating the effect of exercise rehabilitation. The technical solution is as follows:

[0005] The following steps are involved:

[0006] S1: The data acquisition unit collects data during the patient's movement through multi-source sensor fusion, and through the fusion of multi-source data, it can fully reflect the rehabilitation patient's movement status;

[0007] S2: During the multi-source data collection process, the patient's muscle electrical activity signals during exercise are collected through bioelectric signals to understand the degree of muscle activation and fatigue during rehabilitation exercises;

[0008] S3: Develop a rehabilitation assessment model based on the collected multi-source fusion data and muscle electrical activity signal data collected by bioelectric signals;

[0009] S4: Generate a quantitative rehabilitation effect evaluation report based on the output of the evaluation model. The report content includes the values ​​of various quantitative indicators, comparative analysis with rehabilitation goals, and comprehensive score of rehabilitation effect.

[0010] Furthermore, the present application also proposes that multi-source sensor fusion acquisition uses various types of sensors such as accelerometers, gyroscopes, and pressure sensors to collect data such as limb acceleration, angle changes, and joint pressure during the rehabilitation process.

[0011] Furthermore, the present application also proposes to calculate the quantification of rehabilitation effectiveness based on the sports rehabilitation scenario, including the following calculation formula:

[0012] Three-dimensional velocity integral after gravity compensation:

[0013]

[0014] The acceleration in the global coordinate system has gravity removed;

[0015] The joint pressure weighting factor suppresses the noise integration during the unloaded phase;

[0016] When the joint pressure is detected to be lower than the threshold and the gyroscope angular velocity is close to zero, a velocity reset is triggered:

[0017]

[0018] Average joint pressure:

[0019]

[0020] It is obtained by performing differential operations on acceleration data and is used to evaluate the smoothness during movement. The movement path is determined based on the collected data, compared with the standard movement path, and the path deviation index is calculated to quantify the accuracy of the rehabilitator's movement path.

[0021] Furthermore, this application also proposes that the rehabilitation assessment model establishes kinematic indicators, constructs quantitative indicators such as movement speed, acceleration change rate, and movement path deviation, and at the same time establishes functional recovery indicators to establish functional recovery quantitative indicators related to rehabilitation goals.

[0022] Furthermore, this application also proposes that multi-source sensor fusion is collected into wearable constant sensors, and wearable sensor node layout (such as waist, knees, wrists, and limb joints) supports Bluetooth 5.0 transmission.

[0023] Furthermore, the present application also proposes to pre-process the collected data, filter and denoise the data, and extract features (peak torque, standard deviation of joint range of motion).

[0024] A system for quantitative evaluation of sports rehabilitation effects, including a data acquisition system and a rehabilitation evaluation module;

[0025] The data acquisition system includes a multi-source sensor fusion acquisition unit and a bioelectric signal acquisition unit. The multi-source sensor fusion acquisition unit collects kinematic data (joint angles, gait cycles), physiological data (heart rate, electromyographic signals) and patient self-assessment scales through wearable sensors (IMU, electromyoelectric sensors) and visual capture systems (RGB-D cameras). The bioelectric signal acquisition unit uses electromyographic sensors to collect muscle electrical activity signals to understand the degree of muscle activation and fatigue during rehabilitation exercises. For patients undergoing rehabilitation due to muscle injuries, electromyographic signals can intuitively reflect muscle recovery, provide bioelectric data support for quantitative evaluation, and provide directions for constructing quantitative indicators.

[0026] Furthermore, the present application also proposes that the rehabilitation assessment module includes multiple data input units, the multiple data input units are electrically connected to the data comprehensive analysis unit, the data comprehensive analysis unit analyzes and pre-processes the data input from multiple sources, and extracts data features at the same time, the data comprehensive analysis unit is electrically connected to the model reasoning unit, and analyzes and infers the input according to the kinematic indicators and functional recovery indicators of the rehabilitation assessment model components, and finally the conclusions of the analysis and reasoning are comprehensively scored through the evaluation management unit.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] From the above, it can be seen that the present application provides a method and system for quantitative evaluation of sports rehabilitation effects. By integrating multi-source sensor data and bioelectric signals to construct a rehabilitation evaluation model, an evaluation report containing quantitative indicators, comparative analysis and comprehensive scores is generated, which solves the problems of strong subjectivity and single data of traditional evaluation methods, and has the advantages of objectivity, comprehensiveness and precise quantification. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A diagram showing the steps of the method of the present invention;

[0030] Figure 2 Schematic diagram of the system flow of the present invention. DETAILED DESCRIPTION

[0031] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] In existing technologies, the evaluation of exercise rehabilitation effectiveness has long relied on visual observation by rehabilitation physicians and subjective feedback from patients, resulting in inconsistent evaluation standards and low quantification. Although single sensor technology can obtain some motion data, it cannot fully capture the multi-dimensional motion characteristics of the limbs and muscle bioelectric activity, resulting in one-sided evaluation results. When a rehabilitation center conducted a post-operative knee surgery rehabilitation assessment, the traditional method was unable to accurately measure the changes in joint angles and muscle activation timing during the patient's gait cycle, resulting in significant differences in the judgment of rehabilitation progress among different physicians.

[0033] To address these issues, the R&D team discovered that motor function recovery involves complex biomechanical and physiological parameters, necessitating comprehensive assessment through multi-dimensional data fusion. By studying the collaborative working mechanisms of wearable sensor technology and bioelectrical signal acquisition equipment, they attempted to correlate kinematic parameters with muscle electrophysiological characteristics, ultimately developing an evaluation framework for multimodal data fusion. To address data heterogeneity, they established standardized preprocessing procedures and feature extraction methods to ensure the effective integration of multi-source information.

[0034] like Figure 1 A method for quantitatively evaluating the effect of exercise rehabilitation is shown, comprising the following steps:

[0035] S1: The data acquisition unit collects data during the patient's movement through multi-source sensor fusion, and through the fusion of multi-source data, it can fully reflect the rehabilitation patient's movement status;

[0036] S2: During the multi-source data collection process, the patient's muscle electrical activity signals during exercise are collected through bioelectric signals to understand the degree of muscle activation and fatigue during rehabilitation exercises;

[0037] S3: Develop a rehabilitation assessment model based on the collected multi-source fusion data and muscle electrical activity signal data collected by bioelectric signals;

[0038] S4: Generate a quantitative rehabilitation effect evaluation report based on the output of the evaluation model. The report content includes the values ​​of various quantitative indicators, comparative analysis with rehabilitation goals, and comprehensive score of rehabilitation effect.

[0039] This proposal proposes a technical solution that uses multi-source sensor fusion to collect motion data and simultaneously acquire bioelectric signals, construct a rehabilitation assessment model, and generate a quantitative report. The data acquisition unit uses a combination of accelerometers, gyroscopes, and pressure sensors to obtain limb motion trajectories and mechanical parameters, while myoelectric sensors collect surface muscle electrical signals. After filtering and noise reduction, the multi-source data is input into the assessment model, which performs a comprehensive analysis based on kinematic and functional recovery indicators. The final output is an assessment report containing numerical indicators, target comparisons, and a comprehensive score.

[0040] Multi-source sensor fusion acquisition refers to the combination of sensors using complementary measurement principles, such as using an inertial measurement unit to capture three-dimensional spatial motion trajectories and a pressure sensor to monitor plantar force distribution. Specifically, this can be achieved using a nine-axis IMU module integrating an accelerometer, gyroscope, and magnetometer, deployed at major joints to capture complex motion patterns. Bioelectric signal acquisition focuses on extracting the time and frequency domain features of surface electromyographic signals, such as using differential electrodes to collect action potentials during muscle contraction and uploading signal data in real time via a wireless transmission module.

[0041] Specifically, as patients complete standard rehabilitation exercises, sensors worn at key limb locations record motion parameters in real time, while electromyographic sensors simultaneously collect electrical signals from target muscle groups. The collected raw data is filtered through a Butterworth filter to remove high-frequency noise, and a sliding window method is used to extract time-domain features such as motion velocity and joint angle change rate. The preprocessed multimodal data is then fed into a random forest model trained with clinical data. The model then generates a comprehensive assessment index by weightedly integrating kinematic scores and electromyographic activity scores, encompassing dimensions such as motion smoothness, path accuracy, and muscle activation coordination.

[0042] Compared to existing technologies, traditional single-accelerometer solutions cannot distinguish between active movement and environmental interference. This solution effectively eliminates misjudgments through multi-sensor data fusion. Compared to methods that rely solely on kinematic parameters, the introduction of bioelectric signals can detect potential compensatory movement patterns, avoiding superficial functional assessments. Compared to manual assessment methods, this technical solution achieves standardized assessment processes and quantifiable results, addressing the subjective bias in judging rehabilitation effectiveness.

[0043] Through the above technical solutions, this application can objectively quantify the degree of recovery of patients' motor function and accurately identify subtle rehabilitation progress. The multi-source data fusion mechanism effectively improves the integrity of the representation of the movement state, and the collaborative analysis of bioelectric signals and kinematic parameters can reveal the collaborative working state of the musculoskeletal system. The evaluation model realizes the intelligent fusion of multi-dimensional indicators through machine learning algorithms, and the generated quantitative report provides data support for the formulation of personalized rehabilitation plans.

[0044] Multi-source sensor fusion acquisition uses various types of sensors such as accelerometers, gyroscopes, and pressure sensors to collect data such as limb acceleration, angle changes, and joint pressure during the rehabilitation process.

[0045] Among them, the accelerometer refers to a device used to detect the acceleration changes of an object in three-dimensional space. Specifically, it can be implemented using a three-axis MEMS accelerometer. By measuring the acceleration components of the limbs in the X, Y, and Z axes during movement, it provides basic data for calculating the movement speed and acceleration change rate.

[0046] Among them, the gyroscope refers to a device used to measure the angular velocity of an object. Specifically, it can be implemented using a micro-electromechanical system gyroscope. By capturing the angular velocity of rotation around three coordinate axes during limb movement, it is used to calculate the change in joint angle and the continuity index of the motion trajectory.

[0047] Among them, the pressure sensor refers to a device used to detect the pressure distribution of the contact surface. Specifically, it can be implemented using a flexible film pressure sensor. By attaching it to the joint part, it measures the force distribution state of the joint during movement, providing biomechanical data support for evaluating the smoothness of movement.

[0048] Specifically, during exercise rehabilitation, accelerometers collect real-time acceleration data of the limbs in three-dimensional space, integrate the velocity components, and synthesize the total velocity index. Gyroscopes simultaneously measure the angular velocity of limb rotation, convert it into joint angle changes through integration, and use it to calculate the standard deviation index of joint range of motion. Pressure sensors monitor the pressure distribution on the joint contact surface and, combined with the acceleration rate of change index, jointly assess the smoothness of the movement process. By vector-synthesizing the acceleration data and gyroscope data in a spatial coordinate system, a three-dimensional motion trajectory model can be reconstructed. The deviation from the preset standard motion path is then calculated to generate a quantitative index of motion path accuracy.

[0049] This solution addresses the incomplete assessment problem caused by the insufficient dimensionality of single sensor data, enabling simultaneous quantitative assessment of limb acceleration, joint range of motion, and motion smoothness. The spatial fusion of acceleration and gyroscope data allows for precise calculation of joint range of motion standard deviations, while pressure sensor data provides validation of motion load distribution. This combination of data can detect sudden movement deviations that are difficult to capture with traditional methods, thereby improving the objectivity and accuracy of rehabilitation assessments.

[0050] The rehabilitation effectiveness quantification is calculated based on the sports rehabilitation scenario, including the following calculation formula:

[0051] Three-dimensional velocity integral after gravity compensation:

[0052]

[0053] The acceleration in the global coordinate system has gravity removed;

[0054] The joint pressure weighting factor suppresses the noise integration during the unloaded phase;

[0055] When the joint pressure is detected to be lower than the threshold and the gyroscope angular velocity is close to zero, a velocity reset is triggered:

[0056]

[0057] Average joint pressure:

[0058]

[0059] Knee flexion angle (the larger the angle, the less efficient the movement).

[0060] Among them, vx, vy, and vz are the velocity components obtained by integrating the acceleration in the directions of the three coordinate axes. The acceleration change rate index is calculated by performing a differential operation on the acceleration data. It is used to evaluate the smoothness of the movement process. The movement path is determined based on the collected data and compared with the standard movement path. The path deviation index is calculated to quantify the accuracy of the rehabilitator's movement path;

[0061] The motion velocity index refers to the parameter of the velocity vector modulus obtained by integrating acceleration within a three-dimensional coordinate system. Specifically, this can be achieved by using an accelerometer to continuously collect data along three orthogonal axes, then calculating the velocity components of each axis using a numerical integration algorithm, and then synthesizing the vector modulus. This index is used to overcome the limitations of single-direction velocity measurement and reflect the overall changes in limb movement velocity. The acceleration rate of change index refers to the magnitude of the change in acceleration per unit time. Specifically, a differential algorithm can be used to perform a first-order difference operation on the acceleration time series to capture instantaneous fluctuations in acceleration. This index is used to quantify motion smoothness and identify abnormalities in movement continuity. The path deviation index refers to the spatial difference parameter between the actual motion trajectory and the preset standard path. Specifically, the motion trajectory can be solved using gyroscope data and the Euclidean distance or curvature difference can be calculated using a spatial coordinate comparison algorithm. This index is used to measure the accuracy of movement execution and evaluate the rehabilitator's movement control ability.

[0062] Specifically, the acceleration sensor continuously collects acceleration data in three-dimensional space, generates each axial velocity component through integration operation, and obtains the overall motion speed index by synthesizing the velocity vector modulus. For example, in knee flexion and extension rehabilitation training, vx, vy, and vz correspond to the velocity components of the sagittal plane, coronal plane, and horizontal plane respectively, and their modulus v reflects the three-dimensional motion speed of the limb. After the acceleration data is differentially calculated, the acceleration difference between adjacent time points is obtained to form a time series of acceleration change rate. For example, in gait training, this indicator can detect the sudden change in acceleration at the moment of foot contact and evaluate the gait stability. The motion trajectory calculated by the gyroscope data is compared with the spatial coordinates of the standard rehabilitation action path. For example, for the upper limb abduction movement trajectory, the path deviation index is generated by calculating the trajectory curvature difference or the displacement deviation of the key points.

[0063] Compared to existing technologies, traditional methods measure only a single motion parameter using a single sensor, such as using only an accelerometer to measure sagittal plane velocity, making it impossible to comprehensively assess three-dimensional motion velocity. This solution, however, fuses multi-source data, combining three-dimensional accelerometer data with gyroscope trajectory data to construct three indicators: velocity, rate of change of acceleration, and path deviation, forming a multidimensional evaluation system. While existing technologies use acceleration differential processing only to detect the starting point of a movement, this solution expands this into a quantitative indicator of smoothness, enhancing the evaluation dimension.

[0064] This application addresses the lack of quantitative indicators in traditional assessments due to insufficient data dimensionality. It uses three-dimensional velocity synthesis to reflect the overall motion state of the limb, acceleration rate of change to detect instantaneous motion anomalies, and path deviation to quantify motion execution accuracy, providing multi-dimensional quantitative data support for rehabilitation assessment. This multi-dimensional indicator combination can identify complex motion defects that cannot be detected by a single indicator. For example, in gait training, it can simultaneously detect insufficient speed, sudden acceleration changes within the gait cycle, and foot trajectory deviation, improving the comprehensiveness and accuracy of assessment results.

[0065] The rehabilitation assessment model establishes kinematic indicators, constructs quantitative indicators such as movement speed, acceleration change rate, and movement path deviation, and also establishes functional recovery indicators to establish quantitative indicators of functional recovery related to rehabilitation goals;

[0066] Kinematic indicators refer to the physical characteristic parameters of motion calculated through sensor data. Specifically, they can be achieved by performing integral operations and path comparisons on the spatiotemporal data collected by accelerometers and gyroscopes, and are used to quantify the dynamic characteristics of limb movement. Functional recovery indicators refer to clinical recovery parameters set according to individual rehabilitation needs. Specifically, they can be achieved by standardizing data from joint range of motion tests, balance ability tests, or muscle strength tests, and are used to convert subjective rehabilitation goals into measurable objective parameters. The motion speed indicator is the modulus of the three-dimensional space velocity vector obtained by integrating the acceleration data. Specifically, it can be achieved by processing each axial velocity component using a square root operation, and is used to reflect the overall dynamic level of limb movement. The acceleration change rate indicator refers to the differential value of the acceleration data per unit time. Specifically, it can be achieved by processing the original acceleration data using a discrete time series differential operation, and is used to identify sudden movements or abnormal jitters during movement.

[0067] Specifically, during rehabilitation training, wearable sensors continuously collect joint angle and limb movement data, and the raw data is filtered and denoised before being input into the evaluation model. The kinematic index calculation unit calculates the modulus of the motion velocity vector in real time based on the acceleration and gyroscope data as an overall motion intensity index, and at the same time obtains the acceleration change rate through differential operation to reflect the smoothness of the motion. The motion path reconstruction unit uses sensor fusion data to establish a three-dimensional motion trajectory, performs spatial comparison with the preset standard rehabilitation action path, and generates a path deviation value as an indicator of movement accuracy. At the functional recovery level, the evaluation model automatically matches rehabilitation goals based on the patient's medical history information. For example, a flexion angle recovery rate index is set for patients after knee surgery, and the recovery progress is calculated by continuously monitoring changes in the range of joint motion. After weighted fusion of the two types of indicator data, a comprehensive score covering the quality of exercise execution and the degree of physiological function recovery is generated.

[0068] Multi-source sensor fusion acquisition is a wearable constant sensor, with wearable sensor node layout (such as waist, knee, wrist, and limb joints), and supports Bluetooth 5.0 transmission;

[0069] Wearable sensors are sensing devices that combine a flexible circuit substrate with an elastic fixing strap. Specifically, they can be implemented using a micromodule with an integrated accelerometer and gyroscope, combined with a breathable fabric strap. By fitting the sensor to key moving parts of the human body, the sensor ensures stable contact with the skin and reduces motion interference.

[0070] Bluetooth 5.0 transmission refers to a wireless communication protocol based on the 2.4GHz frequency band. It can be implemented using a low-power Bluetooth chip that supports adaptive frequency hopping technology. It ensures the stability of synchronous transmission of multi-node sensor data by increasing the transmission rate and data packet capacity.

[0071] Specifically, sensor nodes are placed at the waist to capture trunk tilt angles and rotational movements, at the knees to monitor knee flexion and extension angles and pressure distribution, at the wrists to capture upper limb swing amplitude and acceleration changes, and at the limb joints to track the movement trajectories of each joint. A Bluetooth 5.0 transmission module is embedded in each sensor node, which transmits data from each node to the data processing terminal in real time by establishing a star topology network, avoiding data delays or movement restrictions caused by traditional wired connections. The sensor layout is adapted to the human anatomy, covering the main muscle groups and joints involved in rehabilitation exercises, ensuring the integrity of kinematic data collection.

[0072] The collected data is preprocessed, filtered, denoised, and features are extracted (peak torque, standard deviation of joint range of motion). Filtering refers to eliminating high-frequency noise in the sensor signal, which can be achieved by using a median filter or a Butterworth low-pass filter to suppress signal distortion caused by environmental electromagnetic interference and equipment vibration.

[0073] Denoising refers to reducing motion artifacts and baseline drift of physiological signals. It can be achieved by using wavelet transform or adaptive filtering algorithm to eliminate signal fluctuations caused by shaking of the limbs of the rehabilitator.

[0074] Peak torque refers to the maximum torque value generated by joint muscles during exercise. It can be achieved by collecting raw data from a torque sensor and taking the maximum value, which is used to quantify the maximum muscle force capacity of the rehabilitator.

[0075] The standard deviation of joint range of motion refers to the degree of discreteness of the change in joint angle, which can be achieved by statistically analyzing the standard deviation of the joint angle time series, and is used to reflect the stability of movement control ability.

[0076] like Figure 1 The system shown is used for quantitative evaluation of sports rehabilitation effects, including a data acquisition system and a rehabilitation evaluation module;

[0077] The data acquisition system includes a multi-source sensor fusion acquisition unit and a bioelectric signal acquisition unit. The multi-source sensor fusion acquisition unit uses wearable sensors (IMU, electromyography sensors) and a visual capture system (RGB-D camera) to collect kinematic data (joint angles, gait cycles), physiological data (heart rate, electromyography signals), and patient self-assessment scales. The bioelectric signal acquisition unit uses electromyography sensors to collect muscle electrical activity signals to understand the degree of muscle activation and fatigue during rehabilitation exercises. For patients undergoing rehabilitation due to muscle injuries, electromyography signals can intuitively reflect muscle recovery, providing bioelectrical data support for quantitative evaluation and the direction of quantitative indicator construction.

[0078] The rehabilitation assessment module includes multiple data input units, which are electrically connected to the data comprehensive analysis unit. The data comprehensive analysis unit analyzes and preprocesses the data input from multiple sources and extracts data features at the same time. The data comprehensive analysis unit is electrically connected to the model reasoning unit, and analyzes and infers the input based on the kinematic indicators and functional recovery indicators of the rehabilitation assessment model components. Finally, the conclusions of the analysis and reasoning are comprehensively scored through the assessment management unit.

[0079] The multi-source sensor fusion acquisition unit is a device combination that integrates multiple sensor modalities. Specifically, it uses an IMU sensor to capture 3D motion posture. For example, the MPU-6050 six-axis sensor can measure acceleration and angular velocity. Myoelectric sensors such as the Myo armband can collect surface electromyographic signals. A visual capture system can use a Kinect V2 RGB-D camera to capture 3D motion trajectories. By synchronously collecting multi-dimensional data, this unit overcomes the technical drawback of insufficient data dimensionality from a single sensor.

[0080] A bioelectric signal acquisition unit is a sensor device specifically designed for monitoring muscle electrical activity. It utilizes bipolar surface electrodes and a differential amplifier circuit to acquire microvolt-level electromyographic signals. For example, a Butterworth bandpass filter eliminates baseline drift. By capturing the timing characteristics of muscle activation, this unit provides objective bioelectrical evidence for quantifying muscle recovery.

[0081] The data integration and analysis unit is a computational module that performs multi-source data fusion processing. Specifically, it employs a Kalman filter algorithm to align inertial sensor and visual data in space and time. For example, it uses a sliding window method to extract the standard deviation of joint range of motion as an indicator of motion stability. By integrating heterogeneous data, this unit overcomes the limitations of traditional assessments that rely on isolated analysis of kinematic and physiological parameters.

[0082] Specifically, the system collects limb kinematic parameters in real time through a network of wearable sensors. For example, waist IMU sensors capture changes in trunk inclination angle, and knee pressure sensors monitor joint load distribution. A visual capture system simultaneously records three-dimensional motion trajectories, for example, calculating gait cycle symmetry using a skeleton tracking algorithm. Myoelectric sensors sample the activation signals of the rectus femoris and vastus lateralis muscles at a 1000Hz sampling rate, for example, calculating muscle activation intensity using the RMS value. During data preprocessing, median filtering is used to eliminate motion artifacts, and wavelet transforms are used to extract frequency domain features of the myoelectric signals. During feature extraction, peak torque is calculated to reflect force output, and the standard deviation of joint range of motion is used to assess motor control ability. The model inference unit performs multimodal correlation analysis between kinematic indicators and myoelectric activation indicators, for example, establishing a temporal correlation model between knee joint angle changes and quadriceps femoris activation intensity. A support vector machine algorithm is then used to output a functional recovery score.

[0083] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. Various changes and improvements are possible without departing from the spirit and scope of the present invention, and such changes and improvements fall within the scope of the invention as claimed.

Claims

1. A method for quantitative evaluation of exercise rehabilitation effects, characterized in that: The following steps are involved: S1. The data acquisition unit collects data during the patient's movement through multi-source sensor fusion, and through the fusion of multi-source data, it can fully reflect the rehabilitation patient's movement status; S2. During the multi-source data collection process, the patient's muscle electrical activity signals during exercise are collected through bioelectric signals to understand the degree of muscle activation and fatigue during rehabilitation exercises; S3, a rehabilitation assessment model is developed based on the collected multi-source fusion data and the muscle electrical activity signal data collected by bioelectric signals; S4. Generate a quantitative rehabilitation effect evaluation report based on the output of the evaluation model. The report content includes the values ​​of various quantitative indicators, comparative analysis with rehabilitation goals, and comprehensive rehabilitation effect scores.

2. A method for quantitative evaluation of exercise rehabilitation effects according to claim 1, characterized in that: Multi-source sensor fusion acquisition uses various types of sensors such as accelerometers, gyroscopes, and pressure sensors to collect data such as limb acceleration, angle changes, and joint pressure during the rehabilitation process.

3. A method for quantitative evaluation of exercise rehabilitation effects according to claim 2, characterized in that: The rehabilitation effectiveness quantification is calculated based on the sports rehabilitation scenario, including the following calculation formula: Three-dimensional velocity integral after gravity compensation: The acceleration in the global coordinate system has gravity removed; The joint pressure weighting factor suppresses the noise integration during the unloaded phase; When the joint pressure is detected to be lower than the threshold and the gyroscope angular velocity is close to zero, a velocity reset is triggered: Average joint pressure: It is obtained by performing differential operations on acceleration data and is used to evaluate the smoothness during movement. The movement path is determined based on the collected data, compared with the standard movement path, and the path deviation index is calculated to quantify the accuracy of the rehabilitator's movement path.

4. A method for quantitative evaluation of exercise rehabilitation effect according to claim 1, characterized in that: The rehabilitation assessment model establishes kinematic indicators, constructs quantitative indicators such as movement speed, acceleration change rate, and movement path deviation, and also establishes functional recovery indicators to establish quantitative indicators of functional recovery related to rehabilitation goals.

5. The method for quantitative evaluation of exercise rehabilitation effect according to claim 1, characterized in that: Multi-source sensor fusion acquisition is wearable constant sensor, wearable sensor node layout, and supports Bluetooth 5.0 transmission.

6. A method for quantitative evaluation of exercise rehabilitation effects according to claim 5, characterized in that: Wearable sensors are distributed on the waist, knees, wrists and limb joints.

7. A method for quantitative evaluation of exercise rehabilitation effects according to claim 1, characterized in that: The collected data are preprocessed, filtered, denoised, and feature extracted, including peak torque and standard deviation of joint range of motion.

8. A method for quantitative evaluation of exercise rehabilitation effects according to claim 7, characterized in that: Data filtering eliminates high-frequency noise in sensor signals and is used to suppress signal distortion caused by environmental electromagnetic interference and equipment vibration, while denoising reduces motion artifacts and baseline drift of physiological signals and is used to eliminate signal fluctuations caused by shaking of the rehabilitator's limbs.

9. A system for quantitative evaluation of sports rehabilitation effects, applied to the method for quantitative evaluation of sports rehabilitation effects according to any one of claims 1 to 8, characterized in that: Includes data acquisition system and rehabilitation assessment module; The data acquisition system includes a multi-source sensor fusion acquisition unit and a bioelectric signal acquisition unit. The multi-source sensor fusion acquisition unit collects kinematic data, physiological data and patient self-assessment scales through wearable sensors and visual capture systems. The bioelectric signal acquisition unit uses electromyography sensors to collect muscle electrical activity signals to understand the degree of muscle activation and fatigue status during rehabilitation exercises. For patients undergoing rehabilitation due to muscle injuries, electromyography signals can intuitively reflect the muscle recovery status, provide bioelectric data support for quantitative evaluation, and provide directions for constructing quantitative indicators.

10. The system for quantitative evaluation of exercise rehabilitation effects according to claim 9, characterized in that: The rehabilitation assessment module includes multiple data input units, which are electrically connected to the data comprehensive analysis unit. The data comprehensive analysis unit analyzes and preprocesses the data input from multiple sources and extracts data features at the same time. The data comprehensive analysis unit is electrically connected to the model reasoning unit, and analyzes and infers the input based on the kinematic indicators and functional recovery indicators of the rehabilitation assessment model components. Finally, the conclusions of the analysis and reasoning are comprehensively scored through the assessment management unit.

Images