A device for monitoring muscle tension abnormalities in renal osteodystrophy

Abstract

This application proposes a monitoring device for abnormal muscle tone in renal osteodystrophy, comprising: an execution module for actively following or passively tractioning the limb under test to complete joint movements; a six-dimensional sensing module for real-time acquisition of raw signals from the limb under test during joint movements; a main control module for closed-loop control of the execution parameters of the execution module; and a data processing module for extracting effective signals reflecting the patient's true muscle tone, performing pathological feature extraction and quantitative calculation, and outputting muscle tone assessment indicators. This application acquires three-dimensional force and torque signals during limb movements in patients with renal osteodystrophy using a six-dimensional force / torque sensor. Combined with dynamic gravity compensation and multi-dimensional data processing, it separates the influence of confounding factors such as edema and malnutrition on total resistance, achieving non-invasive, objective, and accurate quantitative monitoring of muscle tone in patients with renal osteodystrophy. This effectively solves the problems of strong subjectivity in traditional assessments, poor adaptability of existing technologies, and inability to eliminate interfering factors.

Description

Technical Field

[0001] This application relates to the field of medical device technology, and more specifically, to a monitoring device for abnormal muscle tone in renal osteodystrophy. Background Technology

[0002] Currently, the clinical assessment of abnormal muscle tone in patients with renal osteodystrophy mainly relies on manual examination by doctors and subjective grading. This approach has drawbacks such as strong subjectivity, inability to quantify, and limited assessment dimensions. It is difficult to objectively reflect the patient's true muscle strength and tone status, which is not conducive to accurate diagnosis and follow-up of rehabilitation effects.

[0003] Existing muscle strength / tone monitoring devices mostly employ invasive methods such as electromechanical probes, or are only suitable for routine rehabilitation populations such as those with stroke or nerve injury, and are not specifically designed for patients with renal osteodystrophy. While six-dimensional force / torque sensing technology has been applied in rehabilitation robots and sports rehabilitation, it is developed for pathological characteristics of high resistance and velocity dependence, which is completely incompatible with the core pathological characteristics of low resistance and non-velocity dependence in patients with renal osteodystrophy. It cannot distinguish between weak force signals and normal relaxation signals, nor can it eliminate interference from common factors in renal disease patients such as edema and malnutrition. Furthermore, the motion parameters, control strategies, and mechanical structures of existing devices are designed for healthy individuals or routine rehabilitation patients, without considering the characteristics of weak muscle strength, low activity tolerance, and limb edema in patients with renal osteodystrophy. This results in insufficient safety, data distortion, and unreliable monitoring results, leading to a long-standing gap in the application of six-dimensional force control technology in the field of muscle tone monitoring for renal osteodystrophy, failing to meet the clinical needs for precise, non-invasive, and objective quantitative assessment. Summary of the Invention

[0004] This application provides a monitoring device for abnormal muscle tone in renal osteodystrophy. The specific solution is as follows:

[0005] A monitoring device for abnormal muscle tone in renal osteodystrophy, comprising: The execution module is used to complete the preset joint movements by actively following or passively tractioning the patient's limb to be tested, and to provide feedback on the real-time motion parameters during the joint movements. The six-dimensional sensing module is coupled to the force transmission path between the execution module and the limb under test. It is used to collect the original signals of the six-dimensional forces and torques involved in the joint movement of the limb under test in real time and to feed back the original signals. The main control module is used to perform closed-loop regulation of the execution parameters of the execution module based on the original signal and real-time motion parameters, using a control strategy adapted to patients with renal osteodystrophy. The data processing module is used to extract effective signals reflecting the patient's true muscle tone from the original signals and real-time motion parameters, and to perform pathological feature extraction and quantification calculation based on the effective signals, and output muscle tone assessment indicators.

[0006] In some specific embodiments, within the data processing module: First, dynamic gravity compensation is performed on the original signal based on the real-time motion parameters to eliminate torque interference caused by the weight of the limb being tested. Then, non-muscle tone-related interferences associated with the patient are removed to obtain an effective signal that reflects the patient's true muscle tone. In conjunction with the real-time motion parameters, the effective signal is subjected to feature extraction and quantification calculation based on the low resistance and non-velocity-dependent pathological characteristics of abnormal muscle tone in renal osteodystrophy, and muscle tone assessment indicators are output.

[0007] In some specific embodiments, the execution module includes a joint driving unit, which is used to output standardized joint motion to provide the data processing module with accurate real-time motion parameters, including real-time motion angle and motion speed parameters. The joint drive unit has a passive traction mode and an active force generation mode. The passive traction mode has a traction force of ≤10N, and the active force generation mode is used for patient-driven movement and to collect torque signals.

[0008] In some specific embodiments, the six-dimensional sensing module is equipped with a strain gauge six-dimensional force sensor, which accurately captures weak mechanical signals in patients with renal osteodystrophy under low muscle strength.

[0009] In some specific embodiments, a confounding factor separation model specific to renal osteodystrophy is used to eliminate non-muscle tone-related interferences caused by edema and malnutrition associated with the patient. The confounding factor separation model has a built-in edema grading viscous resistance sub-model, which can combine the patient's edema grading and body composition parameters to modify the soft tissue mechanical properties and quantitatively separate non-muscle tone-related viscous resistance.

[0010] In some specific embodiments, the control strategy employs admittance control or impedance control, with stiffness parameters ranging from 200-500 N / m and damping parameters ranging from 5-15 N suitable for patients with renal osteodystrophy and low muscle strength tolerance. s / m.

[0011] In some specific embodiments, the data processing module eliminates the interference of the self-weight of the limb under test through a dynamic gravity compensation algorithm; the dynamic gravity compensation algorithm takes the personalized physiological parameters of the limb under test and the real-time motion parameters as input, calculates the limb gravity torque under the current joint posture in real time, and subtracts the limb gravity torque interference from the original signal.

[0012] In some specific embodiments, the muscle tone assessment indicators include basic muscle strength indicators, endurance indicators, and renal osteodystrophy-specific indicators; the renal osteodystrophy-specific indicators are used to accurately identify abnormal muscle tone characteristics specific to renal osteodystrophy, including the proportion of low resistance signals, torque response delay, and resistance torque variation coefficient at multiple speeds.

[0013] In some specific embodiments, the data processing module extracts specific features based on the torque-angle curve to target the low-resistance, velocity-independent characteristics of abnormal muscle tone in renal osteodystrophy. Within the low-amplitude range of the torque angle curve, by identifying the difference between the patient's true muscle tension signal and active relaxation signal, the low resistance amplitude, the proportion of the low-amplitude segment, the torque fluctuation coefficient, and the resistance torque variation coefficient under multiple speeds are extracted. Based on the resistance torque variation coefficient under multiple speeds, it is determined whether it conforms to the pathological characteristics of low resistance and non-velocity dependence of renal osteodystrophy.

[0014] In some specific embodiments, the execution module further includes a body positioning unit; the body positioning unit is used to coaxially position the limb to be tested and the moving end of the joint drive unit to ensure that the joint rotation center, drive axis and force transmission axis of the six-dimensional sensing module are coaxially aligned.

[0015] Beneficial Effects: This application proposes a monitoring device for abnormal muscle tone in renal osteodystrophy. It collects three-dimensional force and torque signals during limb movement in patients with renal osteodystrophy using a six-dimensional force / torque sensor. Combined with dynamic gravity compensation and multi-dimensional data processing, it separates the influence of confounding factors such as edema and malnutrition on total resistance. This achieves non-invasive, objective, and precise quantitative monitoring of muscle tone in patients with renal osteodystrophy, effectively solving problems such as strong subjectivity in traditional assessments, poor adaptability of existing technologies, and inability to eliminate interfering factors. It significantly improves monitoring accuracy and clinical safety, filling the application gap of six-dimensional force control technology in the field of abnormal muscle tone monitoring in renal osteodystrophy, and providing reliable data support for clinical diagnosis and rehabilitation assessment.

[0016] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the monitoring device of this application; Figure 2 This is a schematic diagram illustrating the operating principle of the monitoring device in this application; Figure 3 This is a complete schematic diagram of the execution module of this application; Figure 4 This is a schematic diagram of the closed-loop control process of the main control module to the execution module in this application; Figure 5 This is a schematic diagram of the data processing flow of the data processing module in this application.

[0019] Reference numerals: 1-Execution module; 2-Six-dimensional sensing module; 3-Main control module; 4-Data processing module; 11-Joint drive unit; 12-Position fixation unit. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0021] This application discloses a monitoring device for abnormal muscle tone in renal osteodystrophy, the module schematic diagram of which is shown in the attached diagram. Figure 1 As shown, a complete monitoring process example is attached. Figure 2 As shown, the specific solution is as follows: A monitoring device for abnormal muscle tone in renal osteodystrophy, comprising: The execution module 1 is used to complete the preset joint movement by actively following or passively tractioning the patient's limb to be tested, and to provide feedback on the real-time motion parameters during the joint movement process. The six-dimensional sensing module 2 is coupled to the force transmission path between the execution module 1 and the limb under test. It is used to collect the original signals of the six-dimensional forces and torques involved in the joint movement of the limb under test in real time and to feed back the original signals. The main control module 3 is used to perform closed-loop control of the execution parameters of the execution module 1 based on the original signal and real-time motion parameters and a control strategy adapted to patients with renal osteodystrophy. Data processing module 4 is used to extract effective signals reflecting the patient's true muscle tension from the raw signals and real-time motion parameters, and to perform pathological feature extraction and quantification calculation based on the effective signals, and output muscle tension assessment indicators.

[0022] The monitoring device in this embodiment mainly consists of an execution module 1, a six-dimensional sensing module 2, a main control module 3, and a data processing module 4. Each module works together to complete the entire process of monitoring, from motion execution, signal acquisition, intelligent control to data processing.

[0023] 101. Enter the relevant pathological parameters of the patient's renal osteodystrophy and the physiological parameters of the limb to be tested, and complete the equipment calibration, test parameter configuration and positioning and fixation of the limb to be tested; 102. The execution module 1, adapted to the low muscle strength characteristics of patients with renal osteodystrophy, drives the limb under test to complete the preset joint movement, and simultaneously collects the real-time motion parameters of the joint. At the same time, the six-dimensional sensing module 2 collects the raw signals of six-dimensional force and torque during the movement of the limb under test in real time. The robot's movement is controlled in a closed loop based on the raw signals and motion parameters collected in real time throughout the process, and a compliant control strategy adapted to the low muscle strength scenario is used to control the robot's movement. 103. Based on the original signal and real-time motion parameters, firstly, signal interference caused by the weight of the limb under test is eliminated based on the real-time motion parameters. Then, through the confounding factor separation model specific to renal osteodystrophy, non-muscle tone-related interference caused by the patient's edema and malnutrition is removed to obtain an effective signal that reflects the patient's true muscle tone. 104. Based on the real-time motion parameters, and considering the low-resistance, non-velocity-dependent pathological characteristics of abnormal muscle tone in renal osteodystrophy, feature extraction and quantification calculation are performed on the effective signals to output a renal muscle tone-specific assessment index. 105. Based on the evaluation indicators, complete the comparative analysis and trend analysis, and automatically generate a standardized monitoring and diagnostic report.

[0024] Execution module 1 is the motion execution component of the device. This module is adapted to the physiological characteristics of patients with renal osteodystrophy, who have weak limb mobility and low exercise tolerance. Through two working modes—active following or passive traction—it drives the patient's limb to complete preset standardized joint movements, with the movement patterns conforming to the routine requirements of clinical muscle strength and tone testing. During the movement, execution module 1 continuously collects and feeds back real-time motion parameters of the joint movement, providing basic motion data for the device's control, adjustment, and data processing. The complete execution module 1 is shown in the attached figure. Figure 3 As shown.

[0025] The six-dimensional sensing module 2 is the mechanical signal acquisition component of the device. This module is coupled in the force transmission path between the execution module 1 and the limb under test, ensuring lossless transmission of mechanical signals. During the movement of the limb driven by the execution module 1, the six-dimensional sensing module 2 non-invasively and in real-time acquires the original six-dimensional force and torque signals generated by the limb under test, completely preserving all mechanical information in the limb movement. It does not employ invasive detection methods such as traditional electromechanical probes, and the entire acquisition process is non-invasive, protecting the patient's physical condition. After acquisition, the original signals are fed back outward.

[0026] The main control module 3 is the core control component of the device. This module receives the raw signals from the six-dimensional sensing module 2 and the real-time motion parameters from the execution module 1. Based on the physiological characteristics of patients with renal osteodystrophy, it selects an appropriate control strategy and performs real-time closed-loop regulation of the execution parameters of the execution module 1, such as the motion speed and amplitude. This ensures that the motion state of the execution module 1 always matches the patient's muscle strength level and tolerance, guaranteeing the safety of the monitoring process and avoiding patient discomfort or data acquisition failure due to mismatched motion parameters. The closed-loop control of the execution module 1 by the main control module 3 is shown in the attached figure. Figure 4 As shown.

[0027] Data processing module 4 is the core analysis component of the device. This module simultaneously receives raw signals and real-time motion parameters, integrates and processes the two types of data, removes invalid interference information, and extracts the effective signals that truly reflect the patient's own muscle tone. Then, combined with the pathological characteristics of abnormal muscle tone in renal osteodystrophy, it performs specific pathological feature extraction and quantification calculations on the effective signals, ultimately outputting standardized muscle tone assessment indicators. This eliminates the subjective judgment method of traditional manual assessment by doctors, achieving intuitive quantification of muscle tone data and providing objective evidence for clinical diagnosis and rehabilitation assessment. The processing flow of data processing module 4 is attached. Figure 5 As shown.

[0028] In some specific embodiments, in the data processing module 4: firstly, dynamic gravity compensation is performed on the original signal based on real-time motion parameters to eliminate torque interference generated by the weight of the limb under test, and then non-muscle tone-related interferences associated with the patient are removed, thereby obtaining an effective signal that reflects the patient's true muscle tone; and combined with real-time motion parameters, the effective signal is feature extracted and quantified based on the low resistance and non-velocity-dependent pathological characteristics of abnormal muscle tone in renal osteodystrophy, and muscle tone assessment indicators are output.

[0029] Data processing module 4 adopts a step-by-step signal processing logic, relying on real-time motion parameters and original mechanical signals to complete the entire process from interference removal to feature quantization, ensuring the authenticity and accuracy of the monitoring results.

[0030] Data processing module 4 first performs preprocessing including noise removal, data smoothing, and outlier removal. Noise removal: a 50Hz notch filter removes power frequency interference, and an 8th-order Butterworth low-pass filter (cutoff 50Hz) removes high-frequency noise, resulting in a signal-to-noise ratio ≥30dB after filtering. Data smoothing: a moving average filter (window of 5-10 sampling points) balances smoothness and response speed. Outlier removal: outliers are replaced with linear interpolation. Then, dynamic gravity compensation is performed based on real-time motion parameters. The principle is that the tested limb generates a constant torque component due to its own weight during joint movement. This component is not the effective torque generated by the patient's active muscle contraction. Since patients with renal osteodystrophy have weak muscle strength, the torque generated by the limb's own weight significantly masks the true muscle strength signal, leading to distortion of the original measurement data. By determining the current motion posture and spatial angle of the limb through real-time motion parameters, the gravitational torque value of the limb segment under the corresponding posture is accurately calculated, and this value is subtracted from the original six-dimensional force torque signal in real time, eliminating the systematic error caused by the limb's own weight from the root, so that the processed signal retains only the mechanical information generated by the patient's active force.

[0031] After gravity compensation is completed, data processing module 4 continues to eliminate interference from non-muscle tone sources. Patients with renal osteodystrophy often experience edema and malnutrition. Edema increases viscous resistance to limb movement, while malnutrition alters the biomechanical properties of soft tissues. The resistance generated by these two factors does not fall under the category of muscle tone and directly interferes with the accuracy of the test results. By separating and removing the additional resistance components caused by the aforementioned pathological factors based on the patient's physiological state, this step yields a true and effective signal that completely eliminates self-weight and pathological interference and reflects only the patient's own muscle contraction ability.

[0032] Subsequently, data processing module 4 combines real-time motion parameters to extract and quantify pathological features. Renal osteodystrophy's abnormal muscle tone is characterized by low resistance and non-velocity dependence, fundamentally different from the biomechanical characteristics of conventional rehabilitation scenarios. Accurate identification requires the integration of parameters such as motion angle and speed. By relying on real-time motion parameters to pinpoint the signal analysis range, specialized analysis is performed on low-amplitude torque signals to distinguish between the patient's weak active exertion and natural relaxation signals, ultimately completing the quantification of various indicators. This processing method effectively solves the problem that conventional signal analysis cannot adapt to the pathological characteristics of renal osteodystrophy. The output muscle tone assessment indicators possess high accuracy and specificity, providing a reliable quantitative basis for clinical diagnosis and rehabilitation assessment.

[0033] In some specific embodiments, a confounding factor separation model specific to renal osteodystrophy is used to eliminate non-muscle tone-related interferences caused by edema and malnutrition associated with the patient. The confounding factor separation model has a built-in edema grading viscous resistance sub-model, which can combine the patient's edema grading and body composition parameters to modify the soft tissue mechanical properties and quantitatively separate non-muscle tone-related viscous resistance.

[0034] Patients with renal osteodystrophy commonly experience edema and malnutrition. Edema increases the viscosity between soft tissues in the limbs, generating additional viscous resistance during movement. Malnutrition alters the mechanical properties of soft tissues such as fat and muscle in the limbs. The resistance generated by these two factors is not part of the muscle tension formed by active muscle contraction and is directly superimposed on the mechanical detection signal, masking the true muscle strength and tension data and failing to reflect the patient's actual muscle function. The confounding factor separation model achieves precise interference separation through a built-in edema grading viscous resistance sub-model. First, the patient's edema grading results and body composition parameters are entered. Then, based on the limb tissue viscosity characteristics corresponding to different edema levels, combined with the soft tissue mechanical parameters reflected by body composition, the resistance component in the signal is modeled and corrected. This quantitatively separates the non-muscle tension-related viscous resistance caused by edema and malnutrition from the total mechanical signal, distinguishing between the true muscle tension signal and pathological interference signals.

[0035] The confounding factor separation model is a multivariate parameterized resistance separation model specifically designed for renal osteodystrophy. Its core is a built-in sub-model of edema grading viscous resistance. This sub-model is a linear fitting correction model based on pathological characteristics. The model inputs include patient edema grade, BMI, body fat percentage, and other body composition parameters. First, a corresponding viscous resistance database is established according to edema grade. Then, the soft tissue mechanical coefficients are corrected by combining body composition parameters. The total non-muscle tone-related viscous resistance caused by edema and malnutrition is calculated through linear fitting. Finally, this resistance is quantitatively separated from the gravity-compensated signal. For example, if a patient has grade III edema and a BMI below the normal range, the model uses the grade III edema viscous resistance coefficient, combines it with body fat parameters to calculate and remove pathological interference resistance values, ultimately obtaining a pure signal that only reflects the patient's true muscle tone.

[0036] In some specific embodiments, the data processing module 4 eliminates the interference of the self-weight of the limb under test through a dynamic gravity compensation algorithm. The dynamic gravity compensation algorithm takes the personalized physiological parameters and real-time motion parameters of the limb under test as input, calculates the limb gravity torque under the current joint posture in real time, and subtracts the limb gravity torque interference from the original signal.

[0037] During joint movement, the mass of the limb being tested will generate a corresponding gravitational torque as the joint posture changes. This torque is generated by the limb's own weight and is not the effective muscle force torque formed by the patient's active muscle contraction. Since the muscle force signal of patients with renal osteodystrophy is relatively weak, the gravitational torque generated by the limb's own weight will significantly mask the true muscle force signal, directly causing the test data to deviate from the actual value. Therefore, it is necessary to use an algorithm to remove this interference in real time.

[0038] Based on classical mechanical torque calculation rules and combined with real-time changes in joint posture, the algorithm dynamically calculates the gravitational torque value of the limb under test in the current state. This value is then subtracted in real-time from the raw mechanical signal acquired by the six-dimensional sensing module 2, achieving precise removal of gravitational interference. This dynamic gravity compensation algorithm can adapt to the physiological differences of different patients' limbs, avoiding errors caused by fixed compensation parameters. Simultaneously, it adjusts the compensation value in real-time following joint movement, eliminating the influence of dynamic changes in gravitational torque on the signal during movement. The compensated raw signal completely eliminates the interference of limb weight, retaining only the true mechanical signal generated by the patient's active muscle contraction.

[0039] The dynamic gravity compensation algorithm belongs to the real-time gravity torque calculation and signal subtraction closed-loop correction algorithm. It is based on local limb biomechanical modeling and relies on classical mechanical torque calculation formulas to complete dynamic calculations. The algorithm inputs are the personalized physiological parameters and real-time motion parameters of the limb under test. The personalized physiological parameters include the mass of the local limb segment, the position of the center of mass, and the coordinates of the joint rotation center. The real-time motion parameters are the current joint motion angle. The algorithm first determines the limb's spatial posture through coordinate transformation, then calculates the limb's gravity torque in the current posture in real time according to the torque formula, and finally subtracts this gravity torque from the original signal synchronously, achieving real-time elimination of self-weight interference. For example, when monitoring the knee joint, the mass of the lower leg and foot segment is 3 kg, the distance from the center of mass to the knee joint rotation center is 0.2 m, and the joint motion angle is 45°. The algorithm calculates the gravity torque value in real time according to the torque formula and subtracts it from the original total torque signal, ensuring that the signal only retains the torque generated by the patient's active muscle force.

[0040] For example, a weighing sensor with a measurement accuracy of ±50g and a laser rangefinder with a measurement accuracy of ±1mm are used to non-invasively collect limb parameters of the area to be measured. The parameters collected and used are those of a local limb segment within the measured area, not the patient's overall body parameters, thus ensuring the accuracy of gravity compensation calculations. Dynamic gravity compensation uses the mass of the local limb segment, the position of its center of mass, the coordinates of the joint rotation center, and the real-time limb motion angle as core input parameters. The gravity used is the weight of the local limb segment itself, obtained by multiplying the local limb segment's mass by gravitational acceleration. The lever arm used to calculate the gravitational torque is the vertical distance between the limb segment's center of mass and the joint rotation center. The system determines the current limb posture based on the real-time motion angle, combines the above core parameters to calculate the gravitational torque generated by the local limb in the corresponding posture in real time, and subtracts this gravitational torque from the original mechanical signal in real time, completing the dynamic gravity compensation.

[0041] The core of calculating gravitational torque is based on the physical parameters and geometric relationships of the limb segment, following the torque formula in classical mechanics: Gravitational torque = local limb segment mass × gravitational acceleration × lever arm × sinθ Local limb segment mass: only for the limb segment being evaluated (such as forearm + hand, lower leg + foot), not the whole body mass (gravity compensation only deducts the interference of the test limb's own weight). Gravitational acceleration: constant (approximately 9.8 m / s²); Lever arm: The straight-line distance from the center of mass of a limb segment to the center of rotation of the joint (i.e., the geometric distance between the two in space). θ: The angle between the plane of limb movement and the direction of gravity, which changes dynamically with the angle of joint movement and needs to be obtained in real time through the robot encoder.

[0042] When assessing elbow flexion muscle strength, first measure the mass of the forearm and hand (m=2kg) and the distance from the center of mass to the elbow joint rotation center (L=0.15m). Then, use an encoder to obtain the angle θ corresponding to the current elbow flexion angle. The gravitational torque M_g = 2 × 9.8 × 0.15 × sin30° = 1.47 N. m.

[0043] Gravity compensation directly targets the measurement data from a six-dimensional force / torque sensor, with the ultimate goal of "correcting muscle strength / tension assessment results," specifically referring to: Direct compensation target: "Total resistance torque / total force" data collected by the sensor - when measuring, the sensor will simultaneously capture "the torque generated by the patient's active muscle force" and "the gravitational torque generated by the limb's own weight", and gravity compensation needs to deduct the latter from the total data; Indirect service recipients: Referencing the algorithms and final results of the rehabilitation assessment system—the compensated data can accurately reflect the patient's active muscle strength (rather than "false data" that includes limb weight), providing accurate input for subsequent feature extraction and torque curve analysis (such as peak torque and average torque calculation).

[0044] The real-time dynamic compensation process consists of three key steps, forming a closed loop: Initial Stage: Gravity Modeling and Parameter Calibration. First, core parameters (mass of local limb segments, position of the center of mass, and coordinates of the joint rotation center) are obtained through precise measurements. A basic gravity model is then established by combining the gravity acceleration (referencing the gravity modeling logic of industrial robotic arms). At the same time, system calibration is completed to ensure the accuracy of parameter measurements (e.g., by correcting measurement deviations using error calibration methods similar to those used in aerospace equipment).

[0045] Real-time: During dynamic adjustment testing in motion (such as isometric contraction and isokinetic contraction tests), the robot encoder collects the joint motion angles in real time, and the compensation algorithm dynamically calculates the current gravitational torque based on the angles; then, the gravitational torque is synchronously subtracted from the real-time measurement data of the six-dimensional force sensor, realizing a real-time closed loop of "measurement-calculation-subtraction".

[0046] Core logic: The final sensor output data = total measured torque - real-time calculated gravitational torque, ensuring that the output results only reflect the torque generated by the patient's active exertion, rather than the interference of the limb's own weight.

[0047] Preferably, considering the physiological characteristics of limb edema often present in patients with renal osteodystrophy, edema directly alters the mass distribution and center of mass of the local limb, thus affecting the accuracy of gravity compensation. Therefore, this gravity compensation model can adaptively adjust parameters according to the patient's edema grade, and correct key parameters such as the mass and center of mass of the local limb segment in real time, so that the compensation result is consistent with the patient's actual limb state, effectively avoiding compensation deviation caused by edema, and maximally restoring the true torque signal generated by the patient's active muscle contraction, providing an accurate and reliable data foundation for subsequent separation of confounding factors and extraction of pathological features.

[0048] In some specific embodiments, the execution module 1 includes a joint driving unit 11, which outputs standardized joint movements to provide the data processing module 4 with accurate real-time motion parameters, including real-time motion angles and speeds. The joint driving unit 11 has a passive traction mode and an active force-generating mode. The passive traction mode has a traction force ≤10N, and the active force-generating mode is used for patient-driven movement and torque signal acquisition. The complete device is shown in the attached figure. Figure 3 As shown.

[0049] As the core component for motion execution in the device, the joint drive unit 11 primarily performs two core functions: outputting standardized joint movements and collecting and feeding back precise real-time motion parameters. The joint drive unit 11 is adapted to the physiological characteristics of patients with renal osteodystrophy, such as weak muscle strength and low exercise tolerance. It features two working modes—passive traction mode and active force exertion mode—to ensure the safety of the monitoring process and the effectiveness of data acquisition. The two modes can be flexibly switched according to the patient's muscle strength status, adapting to patients with different degrees of renal osteodystrophy and comprehensively covering patients with varying muscle strength levels. This balances monitoring safety and data accuracy, providing reliable motion execution and parameter acquisition support for subsequent precise monitoring.

[0050] The passive traction mode is mainly suitable for patients with extremely weak muscle strength who are unable to actively complete joint movements. In this mode, the joint drive unit 11 actively drives the patient's limb to complete the preset joint movement. The traction force is strictly controlled at ≤10N. This traction force threshold is specifically designed to ensure that the limb can follow the movement while avoiding excessive traction force that could cause strain to the patient's weak muscles and joints. At the same time, it guides the patient to become familiar with the range of joint movement, preparing them for possible subsequent active force monitoring.

[0051] The active force exertion mode is suitable for patients with a certain ability to exert force actively. In this mode, the joint drive unit 11 is in a following state, and the patient actively drives the limb to complete the joint movement. The joint drive unit 11 collects the torque signal generated during the patient's active force exertion in real time, and synchronously feeds back parameters such as movement angle and movement speed to ensure that the patient's real muscle strength state during active force exertion can be accurately captured.

[0052] For example, the joint drive unit 11 is driven by a servo motor and harmonic reducer (transmission accuracy ≤0.01mm). Both the knee and elbow joint monitoring systems are designed with one rotational degree of freedom, an adjustable range of motion from 0-120°, a maximum load ≥20kg, and load fluctuation ≤±0.5kg during movement to avoid affecting measurement accuracy. In isometric contraction testing, the joint angle is locked (e.g., 60° for the knee joint), with position control accuracy ≤±0.5°, achieved through a high-resolution encoder (10000 lines / revolution) and PID algorithm. In isokinetic contraction testing, the angular velocity is adjustable from 5° / s to 180° / s (four levels: 30° / s, 60° / s, etc.), with speed fluctuation ≤±2%. 3.2.3 Action mode design: Passive traction mode with traction force ≤5N, used to guide patients to familiarize themselves with the range of motion; active force generation mode where the patient drives the movement, and torque signals are collected in real time. In some specific embodiments, the six-dimensional sensing module 2 is equipped with a strain gauge-type six-dimensional force sensor, which accurately captures the weak mechanical signals of patients with renal osteodystrophy under low muscle strength. The strain gauge-type six-dimensional force sensor is designed specifically for the physiological characteristics of patients with renal osteodystrophy, namely low muscle strength and weak force exertion.

[0053] The working principle of the strain gauge six-dimensional force sensor is as follows: multiple high-precision strain gauges are attached to the surface of its elastic body. When the patient's limb generates force or torque during movement, the elastic body will undergo slight elastic deformation, causing the surface strain gauges to generate corresponding resistance changes. The sensor converts the resistance changes into collectable and transmittable electrical signals through a signal conversion circuit, thereby realizing the synchronous and real-time acquisition of force signals in the X, Y, and Z coordinate axes, as well as torque signals around the three coordinate axes.

[0054] Considering the weak muscle strength of patients with renal osteodystrophy, the mechanical signals generated by their limb exertion have extremely low amplitudes and are easily masked by external noise or interference signals. The strain gauge-type six-dimensional force sensor in this application adopts a high-precision design with a force resolution ≤0.1N and a torque resolution ≤0.01N. With a sampling frequency ≥1000Hz, the sensor accurately captures minute mechanical changes during weak force exertion by the patient, avoiding signal loss or distortion due to insufficient sensor accuracy. Simultaneously, the sensor employs a lightweight and miniaturized design, coupled within the force transmission path between the execution module 1 and the limb being tested. This does not add extra burden to the patient's limb, and force transmission is lossless, ensuring that the acquired raw signals accurately reflect the actual force exertion state of the patient's limb. It is compatible with both passive traction and active force exertion modes of the execution module 1. Whether it's the weak traction force signal during passive traction or the weak torque signal generated by the patient's own muscle contraction during active force exertion, the sensor can stably and accurately acquire and synchronously feed back to the data processing module 4 and the main control module 3.

[0055] In addition, the strain gauge six-dimensional force sensor has good stability and anti-interference ability, which can effectively resist the influence of external factors such as environmental vibration and wire interference during the monitoring process, further ensuring the purity of the original mechanical signal. This provides reliable raw data support for subsequent dynamic gravity compensation, separation of confounding factors and extraction of pathological features, ensuring that the final output muscle tone assessment index can truly and objectively reflect the patient's muscle tone state, and is suitable for the low-resistance, non-velocity-dependent pathological feature monitoring needs of renal osteodystrophy.

[0056] For example, in knee joint monitoring, the sensor is integrated into an adjustable foot pedal, with the force transmission axis coaxial with the knee joint rotation center by ≤2mm. In elbow joint monitoring, the sensor is embedded in an arc-shaped handle made of medical-grade silicone (Shore hardness 50±5HA). Shielded twisted-pair cable is used for signal transmission, with a cable length of 2-3m and waterproof aviation connectors to prevent liquid intrusion. A national Class I force value standard machine (accuracy 0.01%) is used to apply 5%-100%FS standard force / torque to each axis, establishing an input-output mapping model, which is calibrated every 3 months. The sensor response time and phase difference at different frequencies are tested using an electric vibration table (frequency 0-500Hz, maximum acceleration 50m / s²) and a dynamic force hammer to correct dynamic errors. The DH parameter method is used to establish a multi-coordinate system transformation relationship, verified by a calibration fixture with a positioning error ≤0.1mm, resulting in a transformation error ≤1%.

[0057] In some specific embodiments, the control strategy employs admittance control or impedance control, with stiffness parameters ranging from 200-500 N / m and damping parameters ranging from 5-15 N suitable for patients with renal osteodystrophy and low muscle strength tolerance. Both admittance control and impedance control are based on the core principle of force and motion coordinated response, adapting to the core needs of patients with low muscle strength. The two can be flexibly selected according to the monitoring scenario: Admittance control uses the real-time force signal collected by the six-dimensional sensing module 2 as input, and converts the force signal into motion displacement commands for the execution module 1 through preset admittance parameters, achieving real-time response. When the patient exerts weak force, the execution module 1 can simultaneously make a smooth displacement adjustment, avoiding rigid resistance. Impedance control, on the other hand, uses the real-time motion displacement of the execution module 1 as input, and adjusts the output force through impedance parameters, giving the execution module 1 a certain degree of flexible buffering capacity. Even if the patient's force exertion is unstable, the impedance adjustment can offset the impact caused by force fluctuations, ensuring a smooth and controllable movement process.

[0058] To address the physiological characteristics of patients with renal osteodystrophy, such as low muscle strength and poor tolerance, the core parameters of the control strategy were specifically adapted, with stiffness parameters set at 200-500 N / m and damping parameters set at 5-15 N. The stiffness parameter s / m, this range was determined through multiple clinical adaptation tests. The stiffness parameter reflects the resistance of the actuator 1 to changes in motion displacement. If the stiffness parameter is too high, the actuator 1 will exhibit rigid characteristics, and the patient's slight force will not be able to drive the joint movement, and it is easy to cause muscle strain. If the stiffness parameter is too low, the actuator 1 will lack motion stability and will not be able to output standardized joint movements, which will affect the accuracy of real-time motion parameters and torque signal acquisition. A stiffness range of 200-500 N / m can ensure that the patient's slight force can drive the joint movement, while also ensuring the standardization and stability of the joint movement.

[0059] The damping parameter is used to buffer changes in the movement speed of the actuator 1, reducing mechanical impact during movement. If the damping parameter is too low, the movement speed of the actuator 1 will change too quickly, causing instantaneous impact on the patient's joints, which is not suitable for the tolerance characteristics of patients with low muscle strength. If the damping parameter is too high, the movement of the actuator 1 will be too slow, affecting monitoring efficiency and failing to respond promptly to changes in the patient's active force. (5-15N) The damping range of s / m allows the movement speed of the actuator 1 to change smoothly, which can avoid impact injury and follow the patient's exertion state in real time, adapting to both passive traction and active force exertion working modes.

[0060] In some specific embodiments, muscle tone assessment indicators include baseline muscle strength indicators, endurance indicators, and renal osteodystrophy-specific indicators. The renal osteodystrophy-specific indicators are used to accurately identify specific abnormal muscle tone characteristics characteristic of renal osteodystrophy, including the proportion of low resistance signals, torque response delay, and the coefficient of variation of resistance torque at multiple speeds. The muscle tone assessment indicators construct a three-tiered grading system covering baseline muscle strength, exercise endurance, and renal osteodystrophy-specific characteristics. This system not only covers the core dimensions of routine rehabilitation assessment but also specifically highlights the pathological specificity of renal osteodystrophy, comprehensively, accurately, and objectively reflecting the patient's muscle tone status and providing multi-dimensional, high-value data support for clinical diagnosis and rehabilitation assessment.

[0061] Basic muscle strength indicators, as a fundamental dimension for assessing a patient's core muscle contraction ability, mainly include core parameters such as maximum muscle strength and relative muscle strength. Their core function is to quantify the muscle contraction level of the patient's limb under a single maximum exertion state, reflecting the basic ability of active muscle contraction. This indicator is obtained by calculating the original torque signal when the patient's limb completes the maximum range of motion joint movement in the active exertion mode of Module 1. Data processing module 4 extracts and calibrates the peak torque during the active exertion process, combines it with real-time motion angle conversion to obtain the maximum muscle strength value, and then calculates the relative muscle strength based on personalized parameters such as the patient's limb mass and body composition, eliminating assessment bias caused by individual limb differences.

[0062] Endurance indices are used to assess a patient's sustained muscle exertion capacity and fatigue resistance. These indices primarily include parameters such as endurance duration and torque decay rate. The core principle is that muscle endurance directly reflects the long-term stability of muscle energy metabolism and neural regulation, making it a key indicator for assessing a patient's rehabilitation potential and long-term functional status. Within a preset range of joint movement angles and speeds, the change in torque signal over time is monitored during sustained exertion or passive movement. The time it takes for the torque to decrease from its peak value to a preset threshold is calculated as the endurance duration. Simultaneously, the torque decay rate is obtained by quantifying the magnitude of torque decrease at different time points.

[0063] The specific indicators for renal osteodystrophy are designed to accurately identify the unique abnormal muscle tone characteristics of renal osteodystrophy. The proportion of low-resistance signals is calculated by analyzing the low-amplitude segment of the torque-angle curve through data processing module 4, reflecting the proportion of true muscle tone signals to the total signal when the patient is in a low-force state. Due to their weak muscle strength, patients with renal osteodystrophy have a significantly lower proportion of low resistance compared to healthy individuals, making this a core quantitative indicator for identifying low resistance characteristics. Torque response delay is calculated by measuring the time difference between changes in joint angle and feedback torque during active exertion. Patients with renal osteodystrophy have impaired neuromuscular conduction, resulting in a slow torque response; this parameter accurately reflects the abnormal neuromodulation caused by the pathology. The coefficient of variation of resistance torque at multiple speeds is obtained by controlling the execution module 1 to switch between different movement speeds (5° / s-180° / s), collecting resistance torque signals at different speeds, and calculating their fluctuation coefficient. Muscle tone in renal osteodystrophy exhibits a speed-independent characteristic; the fluctuation amplitude of resistance torque at different speeds is small, and the coefficient of variation is significantly lower than in other types of dystonia. This parameter can effectively distinguish renal osteodystrophy from dystonia caused by other diseases such as stroke and Parkinson's disease, improving the specificity and accuracy of the assessment.

[0064] In some specific embodiments, the data processing module 4 extracts specific features based on the torque-angle curve for the low-resistance, non-velocity-dependent characteristics of abnormal muscle tone in renal osteodystrophy. Within the low-amplitude range of the torque-angle curve, by identifying the difference between the patient's true muscle tone signal and the active relaxation signal, it extracts the low-resistance amplitude, the proportion of the low-amplitude segment, the torque fluctuation coefficient, and the resistance torque variation coefficient under multiple velocities. Based on the resistance torque variation coefficient under multiple velocities, it determines whether it conforms to the pathological characteristics of low-resistance, non-velocity-dependent renal osteodystrophy.

[0065] The curve uses the real-time joint motion angle as the horizontal axis and the effective torque signal collected by the six-dimensional sensing module 2 and separated by gravity compensation and confounding factors as the vertical axis. It can intuitively and continuously reflect the dynamic law of the change of the torque value corresponding to the patient's muscle tension with the joint angle throughout the entire joint movement process. The torque value corresponding to each joint angle can accurately correspond to the patient's muscle contraction state at that time.

[0066] The amplitude of the active relaxation signal remained consistently at a very low level, without significant fluctuations, and did not exhibit regular changes with joint angle, appearing only as a stable baseline signal. In contrast, while the true muscle tone signal also fell within a low amplitude range, it exhibited weak, regular fluctuations with changes in joint angle, with the fluctuation amplitude controlled within a preset threshold range. Furthermore, the fluctuation trend corresponded to changes in joint movement angle. By capturing this difference in fluctuations and their correspondence, the active relaxation signal was eliminated, while the effective signal truly reflecting the patient's muscle tone was retained, ensuring the accuracy of subsequent feature extraction.

[0067] The low-resistance amplitude is obtained by extracting the peak values ​​of all effective torque signals within the low-amplitude range and calculating the average peak value, directly quantifying the patient's baseline muscle tone level under low-force conditions. The proportion of the low-amplitude segment is obtained by calculating the ratio of the joint angle range corresponding to the low-amplitude range to the entire preset joint motion angle range, reflecting the duration of the patient's low-resistance muscle tone state. This proportion is significantly higher in patients with renal osteodystrophy than in healthy individuals and other types of patients with abnormal muscle tone. The torque fluctuation coefficient is obtained by calculating the ratio of the standard deviation to the mean of the effective torque signals within the low-amplitude range, reflecting the patient's muscle tone under low-resistance conditions. For tension stability, patients with renal osteodystrophy typically have a higher torque fluctuation coefficient than healthy individuals due to impaired neuromuscular regulation. Extracting the coefficient of variation of resistance torque at multiple speeds requires the use of the speed control function of execution module 1. By controlling execution module 1 to switch between different joint movement speeds (covering preset gears within the range of 5° / s-180° / s, such as 30° / s, 60° / s, etc.), torque-angle curves at different speeds are plotted. The average resistance torque in the low-amplitude range of each curve is extracted, and then the coefficient of variation of these average values ​​is calculated. This parameter is the core indicator for determining the speed-independent characteristics of renal osteodystrophy.

[0068] In some specific embodiments, the execution module 1 further includes a body positioning unit 12; the body positioning unit 12 is used to coaxially position the limb to be tested and the moving end of the joint driving unit 11, ensuring that the joint rotation center, the driving axis, and the force transmission axis of the six-dimensional sensing module 2 are coaxially aligned. The body positioning unit 12 can perform adjustable positioning and fixation according to the size and joint position of the patient's limb to be tested, ensuring that the patient is comfortable to wear, does not compress the limb, and does not affect the blood circulation of the limb, reliably constraining the limb to be tested in a standardized testing posture, so that the joint rotation center of the limb to be tested itself is strictly coaxially aligned with the driving axis of the joint driving unit 11 and the force transmission axis of the six-dimensional sensing module 2.

[0069] Since this device primarily acquires extremely low amplitude muscle force and torque signals from patients with renal osteodystrophy, any limb displacement or axial misalignment will introduce additional eccentric torque and lateral force. The amplitude of such mechanical interference is often much greater than the patient's actual muscle tension signal, directly causing distortion of the original acquired signal. This leads to systematic deviations in gravity compensation calculations and feature parameter extraction. Therefore, triaxial coaxial alignment is a necessary prerequisite for ensuring accurate acquisition of weak signals. After achieving complete triaxial alignment through the body positioning unit 12, the additional torque and lateral force interference caused by limb eccentric placement and joint center misalignment can be structurally eliminated. This allows the six-dimensional sensing module 2 to acquire only pure axial torque signals consistent with the joint rotation direction, preserving the patient's true muscle tension information to the greatest extent possible.

[0070] To ensure the device's detection accuracy, operational stability, and clinical safety, a comprehensive system verification and iterative optimization scheme is implemented. For performance verification, the six-dimensional sensing module 2 is calibrated using a standard force source to verify whether the sensor's detection deviation is within the preset allowable range. Simulated samples are used to verify the processing accuracy of the dynamic gravity compensation algorithm and the confounding factor separation model. Long-term continuous operation tests are conducted to verify the overall operational stability and failure rate of the device. Repeated tests are performed to verify the consistency of various evaluation indicators. Simultaneously, the device's emergency response performance is verified to ensure that the emergency response speed meets safety requirements, and that the device's electrical safety performance complies with relevant national standards for medical electrical equipment. For clinical verification, a controlled trial is conducted, selecting patients with symptoms and healthy subjects as test subjects. The device and traditional assessment methods are used for simultaneous detection. Comparative analysis verifies the consistency and correlation between the two detection methods. Detection efficiency, user comfort, and the device's diagnostic sensitivity and specificity are also statistically analyzed to verify the device's clinical applicability and detection accuracy. In terms of iterative optimization, based on the performance verification and clinical verification results, the parameters and algorithms of the confounding factor separation model and gravity compensation model are optimized, and the structure of the body position fixation structure and human-computer interaction components is improved based on user comfort feedback. At the same time, a long-term optimization mechanism is established to collect user feedback regularly, update the detection reference database regularly, and gradually expand the limb parts that can be monitored according to clinical needs, so as to continuously improve the detection accuracy and performance of the device.

[0071] For example, for knee joint monitoring, the body positioning unit 12 uses an adjustable seat (height 50-70cm) and multiple fixation straps; for elbow joint monitoring, the body positioning unit 12 uses an angle-adjustable forearm support platform, equipped with upper arm and wrist fixation straps. Additionally, red emergency stop buttons (power cut off within 0.5s) are located on both sides of the device, immediately stopping and retracting 5° in case of overload, triggering a buzzer (1kHz) and red light alarm in abnormal situations.

[0072] Those skilled in the art will understand that the modules described above can be implemented using general-purpose computing systems. They can be centralized on a single computing system or distributed across a network of multiple computing systems. Optionally, they can be implemented using computer-executable program code, allowing them to be stored in a storage system for execution by the computing system. Alternatively, they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0073] Note that the above are merely preferred embodiments and the technical principles employed in this application. Those skilled in the art will understand that this application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of this application, the scope of which is determined by the scope of the appended claims.

[0074] The above disclosures are only a few specific implementation scenarios of this application. However, this application is not limited to these. Any variations that can be conceived by those skilled in the art should fall within the protection scope of this application.

Claims

1. A monitoring device for abnormal muscle tone in renal osteodystrophy, characterized in that, include: The execution module is used to complete the preset joint movements by actively following or passively tractioning the patient's limb to be tested, and to provide feedback on the real-time motion parameters during the joint movement process. A six-dimensional sensing module is coupled to the force transmission path between the execution module and the limb under test. It is used to collect the original signals of the six-dimensional forces and torques involved in the joint movement of the limb under test in real time and to feed back the original signals. The main control module is used to perform closed-loop regulation of the execution parameters of the execution module based on the original signal and real-time motion parameters, using a control strategy adapted to patients with renal osteodystrophy. The data processing module is used to extract effective signals reflecting the patient's true muscle tone from the original signals and real-time motion parameters, and to perform pathological feature extraction and quantification calculation based on the effective signals, and output muscle tone assessment indicators.

2. The monitoring device according to claim 1, characterized in that, In the data processing module: First, dynamic gravity compensation is performed on the original signal based on the real-time motion parameters to eliminate the torque interference generated by the weight of the limb under test. Then, non-muscle tone-related interferences associated with the patient are removed, thereby obtaining an effective signal that reflects the patient's true muscle tone. In conjunction with the real-time motion parameters, targeting the low-resistance, non-velocity-dependent pathological characteristics of abnormal muscle tone in renal osteodystrophy, feature extraction and quantification calculation are performed on the effective signals to output muscle tone assessment indicators.

3. The monitoring device according to claim 1, characterized in that, The execution module includes a joint driving unit, which is used to output standardized joint movements and provide the data processing module with accurate real-time motion parameters, including real-time motion angles and motion speed parameters. The joint drive unit has a passive traction mode and an active force generation mode. The passive traction mode has a traction force of ≤10N, and the active force generation mode is used for patient-driven movement and to collect torque signals.

4. The monitoring device according to claim 1, characterized in that, The six-dimensional sensing module is equipped with a strain gauge six-dimensional force sensor, which accurately captures weak mechanical signals in patients with renal osteodystrophy under low muscle strength.

5. The monitoring device according to claim 1, characterized in that, Using a confounding factor separation model specific to renal osteodystrophy, non-muscle tone-related interferences caused by edema and malnutrition associated with patients are eliminated. The confounding factor separation model has a built-in edema grading viscous resistance sub-model, which can combine the patient's edema grading and body composition parameters to correct soft tissue mechanical properties and quantitatively separate non-muscle tone-related viscous resistance.

6. The monitoring device according to claim 1, characterized in that, The control strategy employs admittance control or impedance control, with stiffness parameters ranging from 200-500 N / m and damping parameters ranging from 5-15 N suitable for patients with renal osteodystrophy and low muscle strength tolerance. s / m.

7. The monitoring device according to claim 2, characterized in that, The data processing module eliminates the interference of the self-weight of the limb under test through a dynamic gravity compensation algorithm. The dynamic gravity compensation algorithm takes the personalized physiological parameters of the limb under test and the real-time motion parameters as inputs, calculates the limb gravity torque in the current joint posture in real time, and subtracts the limb gravity torque interference from the original signal.

8. The monitoring device according to claim 1, characterized in that, The muscle tone assessment indicators include basic muscle strength indicators, endurance indicators, and indicators specific to renal osteodystrophy. The indicators specific to renal osteodystrophy are used to accurately identify abnormal muscle tone characteristics specific to renal osteodystrophy, including the proportion of low resistance signals, torque response delay, and the coefficient of variation of resistance torque at multiple speeds.

9. The monitoring device according to claim 1, characterized in that, The data processing module extracts specific features based on the torque-angle curve to target the low-resistance, velocity-independent characteristics of abnormal muscle tone in renal osteodystrophy. Within the low-amplitude range of the torque angle curve, by identifying the difference between the patient's true muscle tension signal and active relaxation signal, the low resistance amplitude, the proportion of the low-amplitude segment, the torque fluctuation coefficient, and the resistance torque variation coefficient under multiple speeds are extracted. Based on the resistance torque variation coefficient under multiple speeds, it is determined whether it conforms to the pathological characteristics of low resistance and non-velocity dependence of renal osteodystrophy.

10. The monitoring device according to claim 3, characterized in that, The execution module also includes a body positioning unit; the body positioning unit is used to coaxially position the limb to be tested and the moving end of the joint drive unit to ensure that the joint rotation center, drive axis and force transmission axis of the six-dimensional sensing module are coaxially aligned.

Images