A post-fracture rehabilitation device, control method and system thereof

Abstract

The application provides a postoperative rehabilitation device for bone fracture, a control method and system thereof, and relates to the technical field of medical instruments. The device comprises: a joint actuator, including a detachable joint module and a driving assembly, the joint module being connected with the driving assembly; a mechanical sensor module; a healthy side data acquisition module, including a force sensor array and an angle measurement unit, the force sensor array being arranged on the surface of the joint actuator in contact with the healthy side limb, and the angle measurement unit being arranged at the joint rotation shaft of the joint actuator; and a controller, the controller being configured to take the mechanical parameters of the healthy side limb as the individualized rehabilitation reference, and dynamically adjust the driving parameters of the driving assembly according to the mechanical data of the affected limb collected by the mechanical sensor in real time. The healthy side data of the patient is a reliable mechanical reference, which can accurately adapt to the individual differences of the patient, and in combination with real-time mechanical feedback, effective step-by-step functional exercise can be carried out, the risk of secondary injury is reduced, and the rehabilitation effect is good.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, and in particular to a post-fracture rehabilitation device, its control method and system. Background Technology

[0002] This invention belongs to the field of medical device technology, specifically relating to an intelligent and personalized rehabilitation device and its design method for postoperative rehabilitation of limb fractures. In particular, it relates to a rehabilitation system based on the biomechanical parameters of the patient's unaffected limb as an individualized rehabilitation benchmark, combining real-time biomechanical feedback, multi-level adaptive adjustment, and remote guidance. This system aims to address the problems of lack of individualized reference, insufficient safety, and poor continuity in traditional rehabilitation training.

[0003] After internal fixation surgery for limb fractures, patients often experience joint stiffness and decreased range of motion due to factors such as pain and prolonged immobilization, which seriously affects their quality of life.

[0004] Currently, the most commonly used rehabilitation devices after fracture surgery are continuous passive motion machines, primarily used for rehabilitation after knee replacement, hip replacement, and limb joint surgery. Their core working principle is: a motor-driven device passively moves the affected limb joints at a constant low speed within a preset angle range, performing flexion and extension movements without the patient needing to exert any effort.

[0005] Currently used rehabilitation training devices are mostly general-purpose, lacking precise adaptation to individual patient differences, leading to poor rehabilitation outcomes and even secondary injuries. Specifically, the devices lack reliable biomechanical references, making it impossible to conduct effective step-by-step functional exercises. Existing rehabilitation devices often set training parameters based on average group data, failing to consider individual differences such as patient age, occupation, and unaffected limb function, resulting in a mismatch between training intensity and the patient's actual tolerance, making it difficult to achieve true step-by-step rehabilitation. Summary of the Invention

[0006] To address the technical problem of existing technologies lacking precise adaptation to individual patient differences, leading to poor rehabilitation outcomes or even secondary injuries, this invention provides a post-fracture rehabilitation device, its control method, and a system. The technical solution is as follows:

[0007] On the one hand, a post-fracture rehabilitation device is provided, the device comprising:

[0008] A joint actuator, comprising a detachable joint module and a drive assembly, wherein the joint module is connected to the drive assembly, and the joint module has a multi-degree-of-freedom motion structure that matches the anatomical movement of human limb joints;

[0009] A mechanical sensor module, comprising a six-axis force sensor, an angle sensor, and an acceleration sensor, wherein the six-axis force sensor is installed at the contact point between the joint actuator and the affected limb, and the angle sensor and acceleration sensor are integrated at the motion joint of the joint actuator;

[0010] The healthy side data acquisition module includes a force sensor array and an angle measurement unit. The force sensor array is disposed on the surface of the joint actuator that contacts the healthy limb, and the angle measurement unit is disposed at the joint axis of the joint actuator.

[0011] The controller has its signal input terminals electrically connected to the mechanical sensor module and the healthy side data acquisition module, respectively, and its control output terminal electrically connected to the drive component. The controller is configured to use the mechanical parameters of the healthy limb as an individualized rehabilitation benchmark, and dynamically adjust the drive parameters of the drive component based on the mechanical data of the affected limb collected in real time by the mechanical sensor.

[0012] Optionally, it also includes a filtering algorithm module;

[0013] Both the mechanical sensor module and the healthy side data acquisition module are connected to the signal input terminal of the controller through the filtering algorithm module.

[0014] Optionally, the joint module of the joint actuator is selected from at least one of the following: hip joint module, knee joint module, ankle joint module, shoulder joint module, wrist joint module, and elbow joint module;

[0015] Each joint module has degrees of freedom matching the anatomical movement of that joint. The knee joint module supports flexion / extension with a single degree of freedom and an angle adjustment range of 0°-145°. The wrist joint module supports flexion / extension with a three-degree-of-freedom compound movement of 0°-90°, radial / ulnar deviation with a range of 0°-30°, and pronation / supination with a range of 0°-180°. The elbow joint module supports flexion / extension with a range of 0°-150° and pronation / supination with a range of 0°-180°. The ankle joint module supports dorsiflexion / plantar flexion with a range of 0°-45° and inversion / eversion with a range of 0°-20°.

[0016] Optionally, the mechanical sensor module further includes a surface electromyography (SEMG) sensor, which is connected to the signal input terminal of the controller;

[0017] The controller includes a compensatory action recognition circuit for recognizing compensatory actions based on signals transmitted by the surface electromyography sensor.

[0018] On the one hand, a control method for a post-fracture rehabilitation device is provided, applicable to any of the above-mentioned post-fracture rehabilitation devices, the method comprising the following steps:

[0019] Step S1: Collect the mechanical parameters of the patient's healthy limb through the healthy side data acquisition module. The mechanical parameters include the isokinetic contraction force, isotonic contraction force, maximum contraction force and maximum range of motion of the healthy limb. The mechanical parameters are then transmitted to the controller as an individualized rehabilitation benchmark for the affected side training.

[0020] Step S2: Set the initial driving parameters for training on the affected side using the controller based on the individualized rehabilitation benchmark. The initial driving parameters include the initial range of motion angles, the initial magnitude of assistive force or resistance, and a safety threshold.

[0021] Step S3: Activate the joint dynamometer, and the drive component drives the affected limb to perform rehabilitation training movements. At the same time, the six-axis force sensor, angle sensor and acceleration sensor in the mechanical sensor module collect the resistance force, joint angle and movement speed data of the affected limb in real time and transmit them to the controller.

[0022] Step S4: The controller compares the real-time collected biomechanical data of the affected limb with the individualized rehabilitation benchmark and safety threshold, and dynamically adjusts the driving parameters of the drive component according to the comparison results, including adjusting at least one of the range of motion angle, the magnitude of assist force or resistance, and the movement speed.

[0023] Step S5: Repeat steps S3 to S4 until the preset training objective is achieved.

[0024] Optionally, the method further includes a compensatory action recognition and correction step:

[0025] During training, at least one of the motion data and electromyographic signals of the affected limb is collected in real time through the mechanical sensor module.

[0026] The controller determines whether there is a compensatory action based on the collected data. The compensatory action includes at least one of trajectory abnormality and compensatory force exertion.

[0027] When a compensatory action is detected, the controller outputs a correction guidance signal.

[0028] Optionally, the step of determining whether a compensatory action exists includes:

[0029] The movement trajectory of the affected limb is collected by angle and accelerometer sensors and compared with the standard trajectory of the healthy side obtained in advance by the healthy side data acquisition module. When the trajectory deviation exceeds the preset angle, it is judged as trajectory abnormality.

[0030] Optionally, the step of determining whether a compensatory action exists includes:

[0031] Electromyography (EMG) signals of the target muscle group and synergistic muscle group in the affected limb are collected by a surface EMG sensor. When the EMG amplitude of the synergistic muscle group or non-target muscle group exceeds the preset ratio of the EMG amplitude of the target muscle group, it is determined to be compensatory exertion.

[0032] Optionally, the method further includes fatigue monitoring and training adjustment steps:

[0033] During training, the controller analyzes the peak force values ​​collected by the six-axis force sensor in real time and calculates the attenuation rate of the peak force values.

[0034] When the attenuation rate exceeds a preset first fatigue threshold, the controller prompts the patient via voice or screen to increase rest time or shorten the duration of a single training session.

[0035] When the attenuation rate exceeds a preset second fatigue threshold, the controller automatically reduces the output resistance or auxiliary force of the drive component, and / or forcibly terminates the current training and generates fatigue recovery suggestions.

[0036] On the one hand, a post-fracture rehabilitation system is provided, including: a cloud server, a patient terminal device, a doctor terminal device, and a post-fracture rehabilitation device as described in any of the above.

[0037] The post-fracture rehabilitation device further includes: a data transmission module, comprising a wireless communication chip and an antenna, wherein the wireless communication chip is electrically connected to the data output terminal of the controller;

[0038] The patient terminal device is equipped with a patient-side application;

[0039] The doctor's terminal device is equipped with a doctor's application;

[0040] The cloud server is communicatively connected to both the patient terminal device and the doctor terminal device, and the data transmission module is communicatively connected to the cloud server via a wireless communication protocol.

[0041] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0042] This device uses the patient's healthy side data as an individualized benchmark to form a reliable biomechanical reference, accurately adapting to individual patient differences. Combined with real-time biomechanical feedback, it can carry out effective step-by-step functional exercises, solving the problem of traditional equipment lacking individualized reference. Closed-loop control enables precise adaptation of training intensity, matching the training intensity with the patient's actual tolerance, avoiding overtraining or undertraining, reducing the risk of secondary injury, and resulting in good rehabilitation effects. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a schematic diagram of the structure of a post-fracture rehabilitation device provided in an embodiment of the present invention;

[0045] Figure 2 This is a flowchart of a post-fracture rehabilitation device control method provided in an embodiment of the present invention;

[0046] Figure 3 This is a working principle and hardware system connection block diagram provided by an embodiment of the present invention;

[0047] Figure 4 This is a mode switching logic diagram provided in an embodiment of the present invention;

[0048] Figure 5 This is a remote collaboration and data flow diagram provided in an embodiment of the present invention.

[0049] Figure label:

[0050] 1. Joint activator; 11. Joint module; 12. Drive assembly; 2. Mechanical sensor module; 3. Healthy side data acquisition module; 4. Controller; 41. Control panel; 5. Support module. Detailed Implementation

[0051] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0052] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0053] In this embodiment of the invention, sometimes a subscript such as W1 may be mistakenly written as a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0054] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0055] This invention provides a post-fracture rehabilitation device. Figure 1 This is a schematic diagram of the structure of a post-fracture rehabilitation device provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the device includes:

[0056] The joint actuator 1 includes a detachable joint module 11 and a drive assembly 12. The joint module 11 is connected to the drive assembly 12. The joint module 11 has a multi-degree-of-freedom motion structure that matches the anatomical movement of the human limb joints.

[0057] The mechanical sensor module 2 includes a six-axis force sensor, an angle sensor, and an acceleration sensor. The six-axis force sensor is installed at the contact point between the joint mover 1 and the affected limb. The angle sensor and the acceleration sensor are integrated at the movement joint of the joint mover 1.

[0058] The healthy side data acquisition module 3 includes a force sensor array and an angle measurement unit. The force sensor array is set on the surface of the joint motion device 1 that contacts the healthy side limb, and the angle measurement unit is set at the joint rotation axis of the joint motion device 1.

[0059] The controller 4 has its signal input terminal electrically connected to the mechanical sensor module 2 and the healthy side data acquisition module 3, respectively. The controller 4 has its control output terminal electrically connected to the drive component 12. The controller 4 is configured to use the mechanical parameters of the healthy limb as the individualized rehabilitation benchmark and dynamically adjust the drive parameters of the drive component 12 based on the mechanical data of the affected limb collected in real time by the mechanical sensor.

[0060] In this invention, the healthy side refers to the limb on the side of the patient that has not suffered a fracture or injury, while the affected side refers to the limb on the side that has suffered a fracture and requires rehabilitation training.

[0061] In this embodiment, the joint mover 1 refers to a mechanical actuator worn on a patient's limb to drive or assist the affected limb in rehabilitation exercises. The joint mover 1 includes a detachable joint module 11 and a drive assembly 12, wherein:

[0062] Joint module 11: It has a multi-degree-of-freedom motion structure that matches the anatomical movement of the joints of the human limbs (including hip joint, knee joint, ankle joint, shoulder joint, wrist joint, and elbow joint), and is connected to the drive component 12. It can be quickly replaced according to the needs of the rehabilitation site.

[0063] Drive component 12: Used to output controllable torque and motion trajectory.

[0064] The Joint Mobility Device 1 features a modular design, with an overall weight controlled between 3-5 kg, making it easy for patients to move and install at home. The Joint Mobility Device 1 is adaptable to different joints of the limbs and offers multi-degree-of-freedom movement capabilities.

[0065] Mechanical sensor module 2 refers to a sensor combination used to collect mechanical and kinematic parameters of the affected limb in real time during rehabilitation training. It is used to collect data on the limb's resistance force, joint angles, and movement speed in real time. It includes:

[0066] Six-axis force sensor: installed at the contact point between the joint actuator 1 and the affected limb, used to simultaneously measure three force components (Fx, Fy, Fz) and three torque components (Mx, My, Mz) in three-dimensional space.

[0067] Angle sensor: Integrated into the moving joint of joint actuator 1, used to monitor the joint movement angle.

[0068] Accelerometer: Integrated into the moving joint of joint actuator 1, used to monitor changes in motion speed and acceleration.

[0069] The healthy side data acquisition module 3 refers to a device used to collect biomechanical parameters of the patient's healthy side limb before rehabilitation training to establish an individualized rehabilitation baseline. It is used to collect data on the patient's healthy side limb's isokinetic contractile force, isotonic contractile force, maximum contractile force, and maximum range of motion before rehabilitation training, serving as an individualized biomechanical reference baseline for affected side training. The healthy side data acquisition module 3 includes:

[0070] Force sensor array: used to collect isokinetic contractile force, isotonic contractile force and maximum contractile force of the healthy limb.

[0071] Angle measurement unit: Located at the joint axis of joint activator 1, used to collect the maximum range of motion of the healthy limb.

[0072] For example, the collected mechanical parameters include: isokinetic contraction force (three speeds: 30° / s, 60° / s, and 90° / s), isotonic contraction force (three loads: 5%, 10%, and 15% of the body weight on the healthy side), maximum contraction force (lasting for 5 seconds), and maximum range of motion (maximum flexion / extension / rotation of the joint).

[0073] Controller 4 refers to a processor with built-in rehabilitation logic algorithms. Its signal input terminals are electrically connected to the mechanical sensor module 2 and the healthy side data acquisition module 3, respectively, and its control output terminal is electrically connected to the drive component 12. Controller 4 has the following functions:

[0074] The control output terminal of the controller 4 is electrically connected to the drive component 12. The controller 4 is configured to use the mechanical parameters of the healthy limb as the individualized rehabilitation benchmark, and dynamically adjust the drive parameters of the drive component 12 based on the mechanical data of the affected limb collected in real time by the mechanical sensor.

[0075] The controller 4 first acquires individualized baseline parameters through the healthy side data acquisition module 3, and establishes an individualized rehabilitation baseline based on the healthy side biomechanical parameters. During training, the controller 4 sets the initial range of motion, assist force magnitude, and safety threshold according to the baseline values.

[0076] After the patient wears the joint movement device 1, the controller 4 controls the drive component 12 to move the affected limb. The mechanical sensor module 2 monitors the actual force and movement status of the affected limb in real time and compares it with the preset safety threshold. When the resistance force, angle, or speed of the affected limb exceeds the allowable range, the controller 4 outputs a control signal to adjust the torque output direction, magnitude, or speed of the drive component 12, realizing closed-loop feedback control to ensure that the training intensity is always within the range that the patient can safely tolerate. The joint movement device 1 supports multi-joint, multi-degree-of-freedom, and detachable rehabilitation exercise training.

[0077] Furthermore, training levels can be set according to the rehabilitation stage, and the training intensity can be adjusted in conjunction with real-time biomechanical feedback.

[0078] For example, controller 4 can use a high-performance embedded processor (such as an ARM Cortex-M7 core) to run rehabilitation control algorithms with a response latency of ≤10ms.

[0079] This device uses the patient's healthy side data as an individualized benchmark to form a reliable biomechanical reference, accurately adapting to individual patient differences. Combined with real-time biomechanical feedback, it can carry out effective step-by-step functional exercises, solving the problem of traditional equipment lacking individualized reference. Closed-loop control enables precise adaptation of training intensity, matching the training intensity with the patient's actual tolerance, avoiding overtraining or undertraining, reducing the risk of secondary injury, and resulting in good rehabilitation effects.

[0080] Specifically, in joint actuator 1:

[0081] The joint module 11 can be connected to the drive assembly 12 via a quick-release interface, enabling faster replacement of different joint modules 11; for example, the replacement of different joint modules 11 can be completed within 3 minutes. The drive assembly 12 includes a low-noise torque motor (output torque range 0-50 N·m) and a precision ball screw transmission structure, with a force control accuracy error ≤ ±0.5 N. Adopting a modular and lightweight design, the overall weight is controlled within 3-5 kg, facilitating home transport and installation for patients. The modules are connected via quick-release interfaces, allowing for the replacement of different joint modules 11 such as the knee, wrist, elbow, and ankle joints within 3 minutes, adapting to the rehabilitation needs of multiple parts of the limbs.

[0082] The joint motion device 1 incorporates a high-precision absolute encoder (resolution ≤0.1°) and a low-noise torque motor (output torque range 0 - 50N・m), combined with a precision ball screw transmission structure, to achieve precise control of the motion trajectory and force control accuracy (error ≤±0.5N), ensuring the effectiveness of rehabilitation training while avoiding additional impact on the joints from mechanical transmission.

[0083] Furthermore, the joint module 11 uses a medical-grade ABS shell and a skin-friendly and breathable liner. The liner is removable and washable, and the shell has a waterproof and dustproof rating of IP54, making it suitable for various home and clinical use scenarios. At the same time, it has passed biocompatibility testing to avoid the risk of skin allergies and infections.

[0084] In mechanical sensor module 2:

[0085] The core uses a six-axis force sensor (range 0 - 100N, 0 - 20N・m) with a sampling frequency of up to 100Hz to collect real-time data on the resistance force and torque changes of the affected limb during training, accurately capturing the details of force exertion and force fluctuations.

[0086] Integrating high-precision angle and acceleration sensors, it synchronously monitors joint movement angle, movement speed (adjustable from 0 to 30° / s), and acceleration changes, constructing a complete motion trajectory data chain to provide a basis for judging the standardization of movements.

[0087] In controller 4:

[0088] Equipped with a high-performance embedded processor and built-in self-developed rehabilitation control algorithm, it has core functions such as real-time data processing, training logic operation, safety threshold judgment and equipment status monitoring, with a response latency of ≤10ms.

[0089] Furthermore, the device also includes a control panel 41, which can be a high-definition touchscreen with an icon-based and text-based user interface. The interface layout is simple and intuitive, making it easy for elderly people and patients with limited mobility after surgery to operate. The interface includes functional modules such as training mode selection, parameter settings, data viewing, device status display, and emergency stop. It also supports voice navigation and voice control, further reducing the operational threshold.

[0090] In one embodiment of the present invention, the device further includes a filtering algorithm module;

[0091] Both the mechanical sensor module 2 and the healthy side data acquisition module 3 are connected to the signal input terminal of the controller 4 through the filtering algorithm module.

[0092] The filtering algorithm module refers to the functional unit used to perform digital signal processing on the raw sensing data collected by the mechanical sensor module 2, and to eliminate environmental interference and motion noise.

[0093] The filtering algorithm module adopts a dual filtering architecture that combines Kalman filtering and moving average filtering:

[0094] Kalman filtering: used to perform optimal state estimation on sensor time-series data, remove Gaussian noise and random interference, suitable for real-time filtering in dynamic motion processes, and can effectively suppress high-frequency noise caused by patient limb tremors, equipment vibration, etc.

[0095] Moving average filtering: used to smooth the data after Kalman filtering and further eliminate residual random fluctuations. It adopts a moving average algorithm with a fixed window length, and the window size can be dynamically configured according to the sensor sampling frequency.

[0096] The core functions of the filtering algorithm module include:

[0097] Real-time noise suppression: Noise reduction is applied to the force and torque signals acquired by the six-axis force / torque sensor to eliminate interference introduced by mechanical transmission and environmental vibration, ensuring the accuracy of the resistance force data of the affected limb.

[0098] Motion signal smoothing: The joint angle, angular velocity and acceleration data collected by the angle sensor and accelerometer are smoothed to eliminate quantization errors and instantaneous jumps, ensuring the continuity and stability of the motion trajectory.

[0099] The response delay of the filtering algorithm module is controlled within ≤10ms to meet the real-time requirements of rehabilitation training and ensure the timeliness of safety threshold judgment and feedback control.

[0100] Specifically, the filtering algorithm module can be integrated into the controller 4 as a software module, or it can be a separate digital signal processing chip. This embodiment adopts a two-stage filtering scheme combining Kalman filtering and moving average filtering. The first stage is Kalman filtering, used to eliminate random noise and systematic errors in sensor measurements; the second stage is moving average filtering (window size set to 5 sampling points), used to smooth signal fluctuations. The raw signals collected by the force sensor module 2 (six-axis force sensor, angle sensor, acceleration sensor) and the healthy data acquisition module 3 (force sensor array, angle measurement unit) are first sent to the filtering algorithm module, filtered, and then transmitted to the controller 4.

[0101] During the baseline acquisition phase, the data from the healthy side is filtered to obtain stable and reliable mechanical parameters. During the training phase, the real-time mechanical data of the affected limb is also filtered to eliminate interference from unconscious shaking of the patient and vibration of the equipment, ensuring that the controller 4 makes accurate judgments.

[0102] The raw sensor signal contains environmental noise, human physiological vibrations, and mechanical transmission noise. The filtering algorithm module estimates the true signal value through a mathematical model, filters out high-frequency noise and abnormal spikes, and outputs a smooth and stable signal to the controller 4.

[0103] This module improves the signal-to-noise ratio and stability of sensor data, avoids accidental triggering of safety protection or incorrect adjustment of drive parameters, and enhances system reliability and patient experience.

[0104] In one embodiment of the present invention, the joint module 11 of the joint activator 1 is selected from at least one of the following: hip joint module 11, knee joint module 11, ankle joint module 11, shoulder joint module 11, wrist joint module 11, elbow joint module 11, each module having a degree of freedom matching the anatomical movement of the joint, supporting compound movements such as flexion, extension, rotation, and deflection.

[0105] This embodiment provides six interchangeable joint modules 11, each designed according to human anatomy.

[0106] Each joint module 11 has a degree of freedom that matches the anatomical movement of the joint. The knee joint module 11 supports flexion / extension single-degree-of-freedom movement with an angle adjustment range of 0°-145°. It adopts a single-axis rotation structure, which is suitable for postoperative rehabilitation of fractures such as distal femoral and tibial plateau.

[0107] The wrist joint module 11 supports three-degree-of-freedom compound movements: flexion / extension 0°-90°, radial / ulnar deviation 0°-30°, and pronation / supination 0°-180°. It adopts a universal joint series structure, which is suitable for rehabilitation after distal radius and wrist bone fracture surgery.

[0108] The elbow joint module 11 supports flexion / extension of 0°-150° and pronation / supination of 0°-180° with dual degrees of freedom. It adopts a biaxial orthogonal structure and is suitable for postoperative rehabilitation of supracondylar fractures of the humerus and olecranon fractures of the ulna.

[0109] The ankle joint module 11 supports dual-degree-of-freedom movements of dorsiflexion / plantar flexion 0°-45° and inversion / eversion 0°-20°. It adopts an eccentric axis structure, which is suitable for postoperative rehabilitation of distal tibia and fibula fractures and talus fractures.

[0110] Hip joint module 11: Provides three degrees of freedom of movement: flexion / extension, adduction / abduction, and internal / external rotation.

[0111] Shoulder joint module 11: Provides three degrees of freedom of movement: flexion / extension, abduction / adduction, and internal / external rotation.

[0112] Each module has a built-in high-precision encoder and a mechanical limit structure to prevent movement beyond its range.

[0113] Based on the anatomical and kinematic characteristics of different joints, the degrees of freedom of movement are precisely matched to ensure that rehabilitation movements are highly consistent with the physiological movement trajectory of the human body.

[0114] Depending on the location of the fracture, the patient or doctor selects the appropriate joint module 11, which is then connected to the drive assembly 12 via a quick-release interface. The controller 4 automatically loads the corresponding kinematic model and angle limitation parameters after identifying the module type.

[0115] This embodiment enables a single device to cover the rehabilitation needs of the major joints of the limbs, reducing patient purchase costs and equipment downtime; anatomically matched degrees of freedom of movement ensure that training movements conform to the natural movement patterns of the human body, avoiding joint compensation or injury.

[0116] In one embodiment of the present invention, the mechanical sensor module 2 further includes a surface electromyography (SEMG) sensor, which is connected to the signal input terminal of the controller 4.

[0117] The controller 4 includes a compensatory action recognition circuit for recognizing compensatory actions based on signals transmitted from the surface electromyography sensor.

[0118] Specifically, the surface electromyography (EMG) sensor uses adhesive disposable electrode pads with a sampling frequency of 2000Hz and a common-mode rejection ratio (CMRR) >100dB. The electrode pads are attached to the target muscle group of the affected limb (such as the quadriceps femoris and biceps brachii) and potentially compensating synergistic muscle groups (such as collecting data from the gluteus maximus for hip joint compensation and the elbow muscles for wrist joint compensation). The controller 4 integrates a compensatory movement recognition circuit, which includes a signal conditioning circuit (amplification and filtering), an analog-to-digital converter, and a microprocessor running the EMG feature extraction algorithm.

[0119] By analyzing muscle activation timing, electromyographic amplitude, and synergistic contraction patterns, compensatory movements (such as hip joint compensatory force in knee rehabilitation and excessive elbow joint involvement in wrist rehabilitation) can be accurately identified, providing data support for movement correction.

[0120] Furthermore, the sensor data is processed by a filtering algorithm (Kalman filtering + moving average filtering) to effectively remove environmental interference and motion noise, ensuring data accuracy and stability. Electromyography (EMG) signal preprocessing (optional): The raw EMG signals acquired by the surface EMG sensor are bandpass filtered and subjected to power frequency notch filtering by the filtering algorithm module to extract effective muscle activation features.

[0121] The recognition logic is as follows: During voluntary activity or resistance training, the root mean square value (RMS) of electromyography of the target muscle and the RMS of the compensating muscle are calculated in real time. When the RMS of the compensating muscle exceeds 50% (preset ratio) of the RMS of the target muscle, it is determined to be compensatory exertion. The controller 4 then generates a correction guidance signal.

[0122] As a preferred option, controller 4 can also combine motion trajectory data for comprehensive judgment.

[0123] Surface electromyography (EMG) sensors collect electrophysiological signals during muscle activity, and compensatory movement recognition circuits analyze the activation level and timing of each muscle group. If abnormal activation occurs in non-target muscles, it indicates that the patient is using an incorrect movement pattern to complete training, and intervention is necessary.

[0124] This embodiment provides a quantitative method for detecting compensatory movements, helping patients establish correct movement patterns and avoid the solidification of incorrect movements, which can lead to decreased rehabilitation outcomes or new injuries.

[0125] Optionally, the device also includes a self-test module to monitor motor temperature, sensor connection status, and battery power in real time; and a temperature and humidity sensor to issue an environmental warning when the ambient temperature is <5℃ or >40℃ and the humidity is >80%.

[0126] This embodiment enables a real-time monitoring and correction mechanism for the patient's movement trajectory, force application pattern, and compensatory movements. During training, patients may experience joint injuries, muscle strains, or reinforce incorrect movement patterns due to improper angles or incorrect force application.

[0127] In one embodiment, the controller 4 of the device has five built-in rehabilitation training modes:

[0128] Level 1: Muscle tremor mode, used to prevent blood clots;

[0129] Second gear: Passive assisted joint movement mode, which adjusts the range of motion based on the biomechanical data of the healthy side and in combination with real-time resistance force;

[0130] Level 3: Autonomous joint movement mode, no assistance required, entirely driven by the patient;

[0131] Gear 4: Low-resistance resistance training mode;

[0132] Gear 5: Medium resistance training mode.

[0133] Furthermore, the device also includes a support module 5 for supporting the joint actuator 1 and the controller 4.

[0134] This embodiment also provides a control method for a post-fracture rehabilitation device, applicable to any of the above-mentioned post-fracture rehabilitation devices. Please refer to [link to relevant documentation]. Figure 2 The method includes the following steps:

[0135] 101: The biomechanical parameters of the patient's healthy limb are collected through the healthy side data acquisition module. The biomechanical parameters include the isokinetic contraction force, isotonic contraction force, maximum contraction force and maximum range of motion of the healthy limb. The biomechanical parameters are then transmitted to the controller as an individualized rehabilitation benchmark for the training of the affected side.

[0136] Preferably, data collection is performed 1-3 days before surgery or within 1 week after surgery. During data collection, the patient completes standardized movements according to the device's voice and screen prompts: 3 repetitions of maximum joint flexion / extension / rotation (maximum value taken); isokinetic contraction, performing 3 maximum contractions at three speeds: 30° / s, 60° / s, and 90° / s; isotonic contraction, performing 3 flexion / extension repetitions at three loads: 5%, 10%, and 15% of the body weight on the healthy side; and maximum voluntary contraction, exerting force for 5 seconds, repeated 3 times, and the average value is taken.

[0137] If a patient has functional impairment in their unaffected limbs, the controller allows the import of a database of biomechanical parameters from people of the same age, gender, and occupation into the system. This data can then be combined with the patient's physical condition to make personalized adjustments and establish alternative rehabilitation benchmarks.

[0138] 102: The controller sets the initial driving parameters for training on the affected side based on individualized rehabilitation benchmarks. The initial driving parameters include the initial range of motion angles, the initial magnitude of assistive force or resistance, and the safety threshold.

[0139] The initial driving parameters are determined by the attending physician after developing a personalized rehabilitation plan based on the patient's fracture type, fixation method, surgical details, and overall health condition. The core contents of the plan include: phased training goals, initial training intensity and duration, range of motion (initial value, incremental rate, target value), resistance level (initial value, adjustment rules), exercise speed, number of repetitions and sets, safety threshold, and fatigue warning threshold.

[0140] The controller automatically calculates initial parameters based on the baseline values: Initial active angle range = maximum angle of the healthy side × 50%; Initial assist force (level two) = maximum contraction force of the healthy side × 20%; Safety threshold (level one warning) = maximum contraction force of the healthy side × 30%, level two safety threshold = 40%. These coefficients can be adjusted.

[0141] 103: Start the joint mobilizer, which drives the affected limb to perform rehabilitation training movements. At the same time, the six-axis force sensor, angle sensor and acceleration sensor in the mechanical sensor module collect the resistance force, joint angle and movement speed data of the affected limb in real time and transmit them to the controller.

[0142] 104: The controller compares the real-time collected biomechanical data of the affected limb with the individualized rehabilitation benchmark and safety threshold, and dynamically adjusts the driving parameters of the drive components based on the comparison results, including adjusting at least one of the following: range of motion angle, magnitude of assist force or resistance, and movement speed.

[0143] The controller compares the resistance force with the safety threshold in real time. If the resistance force is ≤ 30% of the maximum force on the healthy side, the angle range is increased by 5° after each training session (not exceeding the angle on the healthy side); if the resistance force is > 30%, the angle increase stops, and the controller slowly (0.5° / s) returns to the safe position. At the same time, if the force application rate is > 5N / s, it is judged as a sudden force application, and the motor output is immediately reduced.

[0144] 105: Repeat steps 103 to 104 until the preset training objective is achieved.

[0145] Specifically, preset training goals include: a single training session lasting 20 minutes by default, or completing a preset number of repetitions (such as 30 flexion and extension repetitions). Once the conditions are met, the device automatically stops and enters relaxation mode (low-frequency vibration for 3 minutes). The next day's training automatically loads the parameters from the previous session, forming a progressive rehabilitation.

[0146] This method uses data from the healthy side as the standard, and achieves closed-loop control of "assessment-training-reassessment" by comparing the performance of the affected limb with the baseline / threshold in real time, ensuring that each training session is within a safe and effective range. It achieves a fully individualized rehabilitation plan; the dynamic adjustment mechanism adapts to the patient's daily recovery progress; and it has high safety, avoiding rehabilitation failure caused by blindly increasing or decreasing the dosage.

[0147] In one embodiment of the present invention, the method further includes a compensatory action recognition and correction step:

[0148] During training, at least one of the motion data and electromyographic signals of the affected limb is collected in real time through the mechanical sensor module.

[0149] The controller determines whether there is a compensatory action based on the collected data. The compensatory action includes at least one of trajectory abnormality and compensatory force.

[0150] When a compensatory action is detected, the controller outputs a correction guidance signal.

[0151] This step runs in parallel during the execution of steps 103-104. The angle and acceleration sensors in the mechanical sensor module collect motion trajectory data, and the surface electromyography (SEMG) sensor collects electromyographic signals (if configured). The controller has built-in compensation judgment logic: if the trajectory deviates from the standard trajectory of the healthy side by more than 10°, or if the amplitude of the compensating muscle's electromyographic signal exceeds 50% of the target muscle's, it is determined to be a compensatory movement.

[0152] The specific forms of corrective guidance signals include: (1) voice prompts, such as "Please note that hip joint compensatory force has been detected. Please adjust your posture"; (2) on-screen animations demonstrating the comparison between correct and incorrect movements; and (3) tactile feedback from the vibration motors within the device's padding. The controller also records the time, type, and severity of compensatory events and uploads them to the cloud for doctors to analyze.

[0153] As a preferred option, if the same type of compensatory movement occurs in three consecutive training sessions, the system will pause the current training, force the playback of a video demonstrating the correct movement, and continue only after the patient confirms.

[0154] By fusing multimodal biomechanical signals, the system can monitor patients' movement quality and muscle activation patterns in real time. Once an erroneous pattern is detected, intervention can be initiated immediately to help patients establish correct neuromuscular control.

[0155] This embodiment significantly reduces the risk of secondary injury caused by incorrect movement; accelerates the establishment of correct movement patterns and improves rehabilitation efficiency; and provides doctors with objective data on compensatory behaviors, facilitating precise guidance.

[0156] The following examples illustrate the methods for determining two types of compensatory actions: trajectory anomalies and compensatory force exertion.

[0157] Track anomaly:

[0158] Furthermore, the steps to determine whether a compensatory action has occurred include:

[0159] The movement trajectory of the affected limb is collected by angle and accelerometer sensors and compared with the standard trajectory of the healthy side obtained in advance by the healthy side data acquisition module. When the trajectory deviation exceeds the preset angle, it is judged as trajectory abnormality.

[0160] This implementation provides a method for quantitatively judging trajectory abnormalities. During the baseline acquisition phase, when the healthy side performs standard flexion / extension / rotation movements, angle sensors and acceleration sensors record joint angle-time curves and angular velocity-time curves, which are then smoothed and stored in the controller memory as the standard trajectory of the healthy side.

[0161] During training on the affected side, the joint angles and angular velocities of the affected limb are collected in real time. The Dynamic Time Warping (DTW) algorithm aligns the trajectory of the affected limb with the standard trajectory of the healthy side on the time axis and calculates the angle deviation at each moment. When the maximum angle deviation exceeds a preset angle (preferably 10°) or the root mean square deviation exceeds 8°, the trajectory is judged to be abnormal.

[0162] The controller can further analyze the direction and pattern of the deviation: if the deviation is mainly manifested as insufficient range of motion of the affected limb, it may be a muscle strength problem; if it is manifested as abnormal swinging or directional deviation, it may be a compensatory movement. This information is recorded for subsequent adjustments to the rehabilitation plan.

[0163] This embodiment utilizes the high similarity of movement trajectories of both sides of the human body, using the trajectory of the healthy side as a template. The difference between the trajectory of the affected limb and the template reflects the standardization of the movement. It provides an objective and quantitative assessment of movement quality; the preset angle threshold is adjustable to adapt to different rehabilitation stages of patients; and it complements electromyographic signals to improve the accuracy of compensation recognition.

[0164] Compensation efforts:

[0165] The steps to determine whether a compensatory action exists include:

[0166] Electromyography (EMG) signals of the target muscle group and synergistic muscle group in the affected limb are collected by a surface EMG sensor. When the EMG amplitude of the synergistic muscle group or non-target muscle group exceeds the preset ratio of the EMG amplitude of the target muscle group, it is determined to be compensatory exertion.

[0167] This implementation provides a quantitative method for judging compensatory force exertion. Taking knee joint rehabilitation as an example: the target muscle is the quadriceps femoris, and common compensatory muscles are hip extensors (gluteus maximus, hamstrings) or trunk muscles. Surface electromyography electrodes are attached to the rectus femoris (target), gluteus maximus, and erector spinae (candidate compensators).

[0168] Signal processing flow: The raw electromyographic signal is bandpass filtered (20-450Hz), rectified, and lowpass filtered (5Hz) before the root mean square amplitude (RMS) is extracted. Before training begins, the patient performs a maximum voluntary contraction, and the maximum RMS value of the target muscle is recorded as a normalization benchmark. During training, the RMS values ​​of the target muscle and the compensating muscle are calculated in real time (expressed as a percentage of the maximum RMS). When the RMS of the compensating muscle exceeds a preset proportion (preferably 50%) of the RMS of the target muscle, it is determined to be compensatory exertion.

[0169] As a preferred option, this preset ratio can be dynamically adjusted according to the recovery stage: it can be set to 30% in the early stage (more sensitive) and 60% in the later stage (allowing for some synergistic contraction).

[0170] Compensatory exertion is essentially the unconscious use of strong muscle groups by patients to perform actions that should be performed by weaker muscles. Electromyography (EMG) signals directly reflect the neural activation level of muscles, and compensation can be accurately identified by comparing the activation intensity of different muscles.

[0171] This embodiment directly measures muscle activity, detecting compensatory tendencies earlier than simple movement trajectories; it can distinguish between true compensation and normal synergistic contractions; and it provides patients with targeted muscle strength training feedback.

[0172] In one embodiment of the present invention, the method further includes a fatigue monitoring and training adjustment step:

[0173] During training, the controller analyzes multiple consecutive force peaks collected by the six-axis force sensor in real time and calculates the attenuation rate of the force peaks.

[0174] When the attenuation rate exceeds the preset first fatigue threshold, the controller prompts the patient via voice or screen to increase rest time or shorten the duration of a single training session.

[0175] When the attenuation rate exceeds the preset second fatigue threshold, the controller automatically reduces the output resistance or auxiliary force of the drive component, and / or forcibly terminates the current training and generates fatigue recovery suggestions.

[0176] This step is particularly important in resistance training modes (gears four and five). The controller creates a circular buffer to store the peak force of the last 5 contractions (read from the six-axis force sensor). Each time a centripetal contraction is completed, the controller records the peak force F_i and then calculates the total decay rate: decay rate = (F_max - F_current) / F_max × 100%, where F_max is the maximum value among the last 5 contractions.

[0177] The preset thresholds are divided into three levels (using specific values ​​as examples, but not limited to these):

[0178] First fatigue threshold: attenuation rate ≥ 10%. The controller issues a voice prompt: "You may be a little tired. We suggest you rest for 30 seconds or reduce the number of repetitions," while displaying a fatigue icon on the screen.

[0179] Second fatigue threshold: attenuation rate ≥ 20%. The controller automatically reduces the resistance of the next training set by 10-20% and prompts: "Muscle fatigue detected, resistance has been automatically reduced, please continue."

[0180] Third fatigue threshold (optional): Attenuation rate ≥ 30%. The controller forcibly terminates the current training and generates a fatigue recovery suggestion: "You have completed this training. Please take a rest and try again in 2 hours." Simultaneously, the fatigue data is uploaded to the doctor's end.

[0181] Furthermore, the first fatigue threshold is 10%, and the corresponding intervention is to issue a rest suggestion prompt; the second fatigue threshold is 20%, and the corresponding intervention is to automatically reduce the output resistance of the drive component; and when the decay rate exceeds 30%, the controller forcibly terminates the current training.

[0182] If, during training, two consecutive peak force values ​​are lower than 80% of the previous value, the system can trigger resistance reduction or termination in advance.

[0183] When muscles are fatigued, their maximum voluntary contractile force decreases significantly. By analyzing the peak change trend of force exertion over multiple consecutive exertions, the degree of fatigue can be objectively quantified, thus avoiding overtraining.

[0184] It prevents muscle strains or joint injuries caused by overtraining; it automatically adjusts the training load to keep the training intensity within the optimal stimulation range; it reduces the inaccuracy of patients' subjective judgment of fatigue and improves safety.

[0185] Please see Figure 3In one specific embodiment, the method includes a progressive five-level rehabilitation training mode:

[0186] Level 1 (Muscle Tremor Mode): Frequency 10-30Hz, amplitude ±2°, used to prevent deep vein thrombosis.

[0187] Second gear (passive assist mode): Based on the biomechanical data of the healthy side, the initial assist force is 20% of the maximum force of the healthy side, and the angle is 50% of the maximum angle of the healthy side, with real-time feedback adjustment of the resistance force.

[0188] Level 3 (Autonomous Activity Mode): No assistance provided; the patient must exert all their own effort.

[0189] Level 4 (low resistance training mode): resistance 5-15N (corresponding to 10%-30% of the maximum force on the healthy side).

[0190] Level 5 (Medium Resistance Training Mode): Resistance 15-30N (corresponding to 30%-60% of the maximum force on the healthy side).

[0191] Each level can be automatically upgraded according to the recovery progress.

[0192] This method follows the rehabilitation principle of "gradual progress and individualized adaptation" and establishes a five-level progressive training mode to cover the full cycle of rehabilitation needs from the early postoperative period to the recovery period. Each level can be automatically switched according to the patient's rehabilitation progress or manually adjusted by the doctor to ensure that the training intensity is precisely matched with the patient's tolerance.

[0193] The following is a detailed description of the progressive five-level rehabilitation training model:

[0194] Gear 1: Muscle Tremor Mode

[0195] It is suitable for 1-2 weeks after surgery (during the inflammation subsidence period). The core objectives are to promote blood circulation and prevent deep vein thrombosis and muscle disuse atrophy.

[0196] Working principle: The torque motor generates low-frequency (10-30Hz) and small-amplitude (±2°) vibrations, which act on the muscle groups of the affected limb, causing the muscles to passively contract and relax. The vibration intensity is adjustable in 3 levels (weak / medium / strong). The system defaults to the initial intensity of weak, which can be automatically adjusted according to the patient's tolerance or manually set by the patient.

[0197] Safety design: The vibration process monitors the skin temperature and blood circulation of the affected limb in real time (indirectly monitored through a built-in infrared sensor). If any abnormality occurs (such as skin temperature exceeding 38°C), the intensity will be automatically reduced or the operation will be stopped. The default training time is 15 minutes, and training can be repeated every 2 hours. The maximum number of training sessions per day is no more than 4 to avoid overstimulation.

[0198] Second gear: Passive support activity mode

[0199] It is suitable for 2-4 weeks after surgery (the fibrous tissue repair period). The core goal is to gradually restore joint mobility and loosen adhesions to lay the foundation for active movement.

[0200] Working principle: Based on the maximum range of motion and maximum isokinetic contraction force of the healthy side joint, the initial range of motion of the affected limb is set (initially 50% of the maximum angle of the healthy side) and the magnitude of the auxiliary force (initially 20% of the maximum contraction force of the healthy side). During training, the torque motor provides continuous and stable auxiliary force, driving the affected limb to move along the preset trajectory. The six-axis force sensor collects the resistance force of the affected limb in real time. If the resistance force is ≤ 30% of the maximum force of the healthy side (safe threshold), the range of motion is gradually increased in each training session (increased by 5° each time, not exceeding the maximum angle of the healthy side). If the resistance force exceeds the safe threshold, the system immediately stops increasing the angle and returns to the current safe position at a slow speed of 0.5° / s, while simultaneously alerting the patient and doctor with sound and light.

[0201] Personalized adaptation: It supports doctors to adjust the safety threshold (range 20% - 40%) and angle increment rate (3° / time - 8° / time) according to the patient's fracture type and fixation method (such as plate fixation, intramedullary nail fixation) to ensure the safety and effectiveness of training.

[0202] Gear 3: Autonomous Activity Mode

[0203] It is suitable for 4-8 weeks after surgery (muscle function recovery period). The core goal is to strengthen the active movement ability of the affected limb, improve muscle control, and reduce compensatory movements.

[0204] Working principle: The device does not provide assistance; the patient drives the joint movement entirely by exerting their own force. The system monitors the movement trajectory, speed, and continuity in real time through angle and acceleration sensors. Surface electromyography (EMG) sensors (optional) simultaneously analyze muscle activation patterns and identify compensatory movements (such as excessive shoulder involvement in elbow rehabilitation and knee flexion compensation in ankle rehabilitation). If improper or compensatory movements occur, the system guides the patient to correct them through voice prompts (such as "Please keep your elbow stable and avoid shoulder exertion") and on-screen animations. At the same time, abnormal movement data is recorded and uploaded to the doctor's end for subsequent guidance and reference.

[0205] Protective design: Set an upper limit (not exceeding 90% of the maximum angle of the healthy side) and a lower limit (not lower than 0°) for the joint's range of motion. When the angle approaches the limit, the system will remind the patient through tactile feedback (slight vibration of the device pad). If there is a sudden excessive force (exceeding 50% of the maximum force of the healthy side), the system will activate emergency braking to prevent excessive force on the joint.

[0206] Training statistics: Automatically records the number of effective movements (movements with standardized movements and within the preset angle range), average movement speed, and force uniformity score for each training session, providing data support for rehabilitation effect evaluation.

[0207] Gear 4: Low Resistance Training Mode

[0208] It is suitable for 8-12 weeks after surgery (muscle strength enhancement period). The core goal is to enhance the muscle strength of the affected limb, improve muscle endurance, and further consolidate joint mobility.

[0209] Working principle: Based on the autonomous activity mode, the device provides adjustable resistance through the built-in magnetorheological damper (resistance range 5-15N, corresponding to 10%-30% of the maximum contraction force of the healthy side); during training, the resistance remains constant, and the patient needs to overcome the resistance to complete joint flexion, extension, rotation and other movements. The system records the force curve and resistance resistance data in real time and analyzes the characteristics of muscle force distribution (such as peak muscle force and duration of force).

[0210] Dynamic adjustment: The resistance is automatically adjusted based on the patient's training data. If the peak force of three consecutive training sessions reaches 120% of the current resistance, the resistance will be automatically increased in the next training session (by 10% of the current resistance). If the peak force is lower than 80% of the current resistance, the current resistance will be maintained or appropriately reduced (by 5% of the current resistance) to ensure that the training intensity is always within the patient's "effective training range".

[0211] Training modes: Supports two modes: isokinetic resistance training (constant exercise speed, resistance varies with force exertion) and isotonic resistance training (constant resistance, exercise speed varies with force exertion). Doctors can choose or switch between modes according to the patient's rehabilitation needs.

[0212] Gear 5: Medium Resistance Training Mode

[0213] It is suitable for patients 12 weeks or more after surgery (functional consolidation period). The core goal is to restore the muscle strength of the affected limb to the level of the healthy side and improve joint function and daily living activities.

[0214] Working principle: The resistance range is 15-30N (corresponding to 30%-60% of the maximum contraction force of the healthy side). It adopts a progressive resistance loading design. In the initial stage of training, the resistance is 30% of the maximum contraction force of the healthy side, and it is gradually increased as the training progresses, with a maximum of no more than 60% of the maximum contraction force of the healthy side. It supports multiple training modes (3 sets by default, 10-15 repetitions per set, with 1-2 minutes of rest between sets). The number of sets, repetitions and rest time can be adjusted by the doctor.

[0215] Muscle strength assessment: Muscle strength assessment is automatically completed during training. By analyzing indicators such as maximum force, force exertion speed, and endurance decay rate, a muscle strength score (0-10 points) is generated and compared with the muscle strength data of the healthy side to intuitively show the degree of muscle strength recovery. At the same time, combined with joint range of motion data, the functional recovery level of the affected limb is assessed, providing a basis for whether to enter the later stage of rehabilitation or stop training.

[0216] Safety restrictions: Set the maximum resistance limit for a single training session (not exceeding 60% of the maximum contraction force of the healthy side) and the maximum number of sets per day (not exceeding 5 sets) to avoid overtraining that could lead to muscle strain or joint injury; if the muscle strength decline rate exceeds 50% during training (indicating fatigue), the system will automatically remind you to rest or forcibly terminate the current training.

[0217] The implementation of this method, based on individualized rehabilitation benchmarks, achieves precise adaptation with a "one person, one policy" approach.

[0218] The innovation uses the biomechanical parameters of the patient's healthy side as the core reference, replacing the traditional group average data. It fully considers individual differences such as age, gender, weight, occupation, muscle strength, and joint function, and establishes a truly individualized rehabilitation benchmark. This ensures that the training intensity, range of motion, and resistance are precisely matched with the patient's actual tolerance, avoiding poor rehabilitation results or secondary injuries caused by a "one-size-fits-all" approach.

[0219] For patients with functional impairment of the unaffected limbs, a personalized correction model based on a population database is constructed. By fitting multi-dimensional data, an alternative benchmark is generated to address the lack of rehabilitation references for special populations and expand the applicability of the equipment.

[0220] 2. Real-time biomechanical feedback and comprehensive safety protection enhance rehabilitation safety.

[0221] Integrating multi-dimensional sensing technologies such as six-axis force sensors, angle sensors, and electromyography sensors, it enables real-time monitoring of resistance force, joint angles, movement trajectories, and muscle activation states. The data sampling frequency is as high as 100Hz - 2000Hz, ensuring that every detail change during training is captured, providing data support for dynamic adjustment and safety protection.

[0222] A five-layer safety protection system is constructed, consisting of "force control protection, angle restriction, compensation identification, fatigue monitoring, and equipment self-inspection." Through mechanisms such as dual-threshold early warning, progressive angle opening, real-time movement correction, and dynamic fatigue adjustment, training risks are avoided from multiple dimensions, minimizing the incidence of complications such as joint injury, muscle strain, and thrombosis, thus addressing the core pain point of insufficient safety in traditional rehabilitation training.

[0223] 3. A multi-level, progressive training system to cover rehabilitation needs throughout the entire cycle.

[0224] The innovative design features five training modes to precisely match the core goals of different postoperative rehabilitation stages: from early postoperative muscle tremors (to prevent thrombosis), to mid-term passive assisted activities (to restore mobility) and voluntary activities (to improve control), and then to late-term low / medium resistance training (to enhance muscle strength), forming a complete rehabilitation ladder. This ensures that the training process is gradual, scientific, and reasonable, avoiding poor results or injuries caused by a mismatch between training intensity and rehabilitation stage.

[0225] The training parameters for each level support fine adjustment (such as resistance level accuracy of 1N, angle adjustment accuracy of 0.1°, and multiple adjustable exercise speeds), and doctors can flexibly switch and adjust them according to the patient's recovery status to meet the personalized rehabilitation needs of different patients, while also adapting to the rehabilitation characteristics of different fracture types and fixation methods.

[0226] This invention also provides a post-fracture rehabilitation system; please refer to [link / reference]. Figure 4 and Figure 5 The system includes: a cloud server, patient terminal equipment, doctor terminal equipment, and post-fracture rehabilitation device as described above;

[0227] The post-fracture rehabilitation device also includes: a data transmission module, including a wireless communication chip and an antenna, wherein the wireless communication chip is electrically connected to the data output terminal of the controller;

[0228] The patient-side terminal device has a patient-side application installed;

[0229] The doctor's terminal device has a doctor-side application installed;

[0230] The cloud server is connected to both the patient's terminal device and the doctor's terminal device, and the data transmission module is connected to the cloud server via a wireless communication protocol.

[0231] The data transmission module is a communication unit used to upload training data to a cloud server or physician terminal in real time. The data transmission module includes a wireless communication chip and antenna, and establishes a data connection via communication protocols such as Bluetooth, Wi-Fi, or wired Ethernet, supporting local data storage in case of network interruption.

[0232] The remote guidance platform refers to a three-in-one rehabilitation data interaction system consisting of a patient terminal, a doctor terminal, and a cloud terminal, including:

[0233] Patient-side terminal device: It has a patient-side application installed, which is used to display personalized training plans, display training data curves in real time, receive doctor guidance messages, and upload training videos.

[0234] Doctor-side terminal device: It has a doctor-side application installed, which is used to view the patient's healthy side baseline data and previous training data, monitor the training process in real time, remotely adjust training parameters and levels, and generate rehabilitation assessment reports.

[0235] Cloud server: It communicates and connects with both the patient and doctor's ends, and has a built-in big data analysis engine and rehabilitation effect prediction model to deeply analyze training data and generate rehabilitation suggestions.

[0236] The system supports three data transmission methods: Bluetooth, wired Ethernet, and more. It can be flexibly switched according to the usage environment to ensure that training data is uploaded to the cloud platform in real time and stably. It also supports local data storage to avoid data loss due to network interruption.

[0237] This system integrates rehabilitation devices with a telemedicine platform. The data transmission module antenna within the rehabilitation device itself is integrated into the device's casing. Every second, the controller packages training data (angle, force, electromyography, fatigue indicators, etc.) and uploads the training data to the cloud or physician's terminal in real time.

[0238] The cloud server is built on Alibaba Cloud or AWS platform, using a MySQL database to store patient files, healthy side baseline data, and all training records. It also deploys a Python-based rehabilitation effect prediction model (such as linear regression or random forest algorithms). Patient and doctor terminals are smartphones or tablets with a dedicated app installed (supporting iOS and Android). The patient app functions include: viewing training plans, real-time data curve display, receiving doctor messages, uploading training videos, and scheduling appointments. The doctor app functions include: viewing healthy side baseline and all training data, remotely adjusting training parameters (level, angle limit, resistance level), generating rehabilitation assessment reports, and providing real-time video guidance.

[0239] The communication connection follows the HTTPS protocol and the WebSocket protocol. The former is used for data uploading and command issuance, while the latter is used for real-time monitoring.

[0240] Furthermore, the controller has a built-in rehabilitation logic algorithm that supports both patient self-control and remote physician guidance modes.

[0241] When patients are training at home, the device automatically uploads data to the cloud. Doctors can view the data at any time through their app and remotely adjust the treatment plan or initiate a video call if any abnormalities are found. Patients receive guidance synchronously through the app, achieving a closed-loop rehabilitation process between the hospital and home.

[0242] The controller allows patients or physicians to customize training parameters, including range of motion, movement speed, resistance, training duration, and number of repetitions.

[0243] The system also includes training data visualization and intelligent analysis functions, which can display training curves, progress reports, and abnormal alerts in real time through a mobile APP or computer, and provide doctors with adjustment suggestions.

[0244] This embodiment constructs dedicated apps for doctors and patients, supports both iOS and Android systems, and also provides a web-based platform for computers to meet the usage habits of different users.

[0245] Patient-side APP functions: Display personalized training plans (including gear level, angle range, resistance level, training duration, number of repetitions, etc.), display training progress and data curves in real time (angle-time, force-time, electromyography signal curves), receive doctor guidance messages and abnormal alerts, upload training videos (supports real-time recording and local upload), view rehabilitation reports and historical data, and schedule remote consultations.

[0246] Doctor-side APP / web version functions: View patient's baseline data and training data for the healthy side, monitor the patient's training process in real time (supports video connection and data synchronization), remotely adjust training parameters and levels, issue training guidance and movement correction suggestions, generate weekly / monthly rehabilitation assessment reports, manage patient files in batches, and set safety thresholds and early warning rules.

[0247] The controller automatically calculates initial parameters based on baseline values, which can be adjusted via the doctor's terminal. During various stages of use, the controller can also send alerts to the doctor's terminal.

[0248] The cloud-based data analysis platform uses big data and artificial intelligence algorithms to perform in-depth analysis of patient training data, including the rate of improvement in activity level, muscle strength recovery trend, training compliance score, and movement standardization assessment. It automatically generates rehabilitation effect prediction models to provide data support for doctors to adjust training programs. It also has data encryption storage function to protect patient privacy.

[0249] Based on this system, patients can receive professional guidance and supervision after discharge, thereby improving rehabilitation compliance, thoroughly implementing the training plan, improving long-term rehabilitation effects, and avoiding joint adhesions.

[0250] Based on this device, it has the functions of real-time data transmission and remote expert guidance, which improves the communication between doctors and patients and allows for timely adjustments to the rehabilitation plan.

[0251] By constructing a three-in-one remote rehabilitation platform integrating "patient-cloud-doctor", real-time transmission of training data, remote distribution of rehabilitation plans, real-time monitoring of the training process, and real-time communication between doctors and patients can be achieved. This breaks through the limitations of time and space, solves the problems of lack of professional guidance, decreased rehabilitation compliance, and incomplete implementation of training plans after patients are discharged, ensures seamless connection between in-hospital and home rehabilitation, and improves the continuity of rehabilitation.

[0252] A cloud-based analytics platform based on big data and artificial intelligence algorithms can deeply mine the value of training data, generate rehabilitation progress reports, effect prediction models, and personalized adjustment suggestions, providing data support for doctors and improving the professionalism and accuracy of rehabilitation guidance; at the same time, it allows patients to intuitively understand their rehabilitation progress, enhancing their confidence and compliance with training.

[0253] Please see Figure 4 and Figure 5 Based on the above-mentioned post-fracture rehabilitation system, the system is also equipped with the following rehabilitation management process:

[0254] I. Baseline Assessment and Rehabilitation Record Establishment

[0255] During the rehabilitation preparation phase, the biomechanical parameters of the patient's healthy limb are acquired through the healthy side data acquisition module. The cloud system preprocesses and analyzes the acquired data, generates a biomechanical parameter report of the healthy limb, establishes a personal rehabilitation file for the patient, and clarifies rehabilitation benchmark values ​​(including maximum safe angle, initial training resistance, and safety threshold), providing a basis for the subsequent rehabilitation plan.

[0256] II. Development and Issuance of Rehabilitation Plan

[0257] Based on the patient's fracture type, fixation method, and baseline assessment data, doctors develop personalized rehabilitation plans using a doctor-side application, including:

[0258] Key parameters: range of motion (initial value, incremental rate, target value), resistance level (initial value, adjustment rules, maximum value), movement speed (adjustable from 30° / s to 90° / s), number of repetitions and sets;

[0259] Safety parameters: safety threshold (20% to 40% of the maximum force on the healthy side), fatigue warning threshold (attenuation rate 10% to 20%), equipment operating parameters (vibration frequency, auxiliary force magnitude).

[0260] Once the plan is developed, the doctor sends it to the patient's application via a cloud server. After receiving the plan and confirming its accuracy, the patient signs an electronic informed consent form, and the device automatically synchronizes the plan parameters.

[0261] III. Training Execution and Remote Supervision

[0262] Preparation before training: The patient completes the device self-check, wears the joint module, and takes a photo of the affected limb's condition through the patient-side application (optional) and uploads it to the doctor's end. After playing the training guidance video, training begins.

[0263] Training Process: Patients complete standardized movements according to voice and screen guidance. The device collects data in real time, including resistance force, joint angle, movement speed, and electromyographic signals, and displays an angle progress bar and force curve on the screen. The system provides real-time feedback and adjustments based on preset rules: it immediately provides audio and visual alerts and corrections for excessive force, exceeding angle limits, or compensatory movements; after reaching the training goal, it automatically enters relaxation mode (mild vibration for 3-5 minutes). Patients can stop training urgently and record the reason via a physical button or a virtual button on the app.

[0264] Remote supervision and guidance:

[0265] Doctors can view patients' training data curves and training videos in real time through a doctor-side application, and provide real-time guidance via text, voice, images, animations, or video calls when abnormalities are detected.

[0266] The system automatically identifies training anomalies (frequent safety warnings, insufficient training time, poor movement standardization) and sends warning reminders to the doctor, who then intervenes promptly.

[0267] Patients can consult with doctors through the app, and doctors will respond within 24 hours (within 1 hour for emergencies).

[0268] IV. Rehabilitation Progress Assessment and Plan Adjustment

[0269] Periodic assessment: The system automatically generates weekly / monthly rehabilitation reports, including:

[0270] Training execution status (number of repetitions, duration, compliance score); rehabilitation effect data (progress in joint range of motion, muscle strength recovery rate, force uniformity, incidence of compensatory movements); safety indicators (number of warning triggers, adverse reactions); rehabilitation prediction (remaining time, key directions for the next stage).

[0271] Plan Adjustment: After receiving the report, the doctor will conduct a comprehensive assessment based on the patient's clinical examination results (such as fracture healing X-rays): If the goal is achieved: adjust the plan for the next stage (increase the intensity, increase the strength, update the safety threshold); if the goal is not achieved: analyze the reasons and make targeted adjustments (strengthen supervision, reduce the intensity, extend the training time); if the progress is slow due to clinical reasons such as poor fracture healing, the training should be adjusted or suspended in conjunction with the clinical treatment plan.

[0272] Determination of rehabilitation endpoint: When the patient's range of motion of the affected limb joint reaches more than 90% of the maximum angle of the unaffected side, the muscle strength recovery rate reaches more than 80%, the ability to perform daily living activities recovers to the pre-operative level, and there are no training abnormalities or adverse reactions for two consecutive weeks, the doctor determines that the rehabilitation endpoint has been reached and issues a rehabilitation completion notification and subsequent maintenance training suggestions through the cloud server.

[0273] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0274] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0275] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0276] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A post-fracture rehabilitation device, characterized in that, The device includes: A joint actuator, comprising a detachable joint module and a drive assembly, wherein the joint module is connected to the drive assembly, and the joint module has a multi-degree-of-freedom motion structure that matches the anatomical movement of human limb joints; A mechanical sensor module, comprising a six-axis force sensor, an angle sensor, and an acceleration sensor, wherein the six-axis force sensor is installed at the contact point between the joint actuator and the affected limb, and the angle sensor and acceleration sensor are integrated at the motion joint of the joint actuator; The healthy side data acquisition module includes a force sensor array and an angle measurement unit. The force sensor array is disposed on the surface of the joint actuator that contacts the healthy limb, and the angle measurement unit is disposed at the joint axis of the joint actuator. The controller has its signal input terminals electrically connected to the mechanical sensor module and the healthy side data acquisition module, respectively, and its control output terminal electrically connected to the drive component. The controller is configured to use the mechanical parameters of the healthy limb as an individualized rehabilitation benchmark, and dynamically adjust the drive parameters of the drive component based on the mechanical data of the affected limb collected in real time by the mechanical sensor.

2. The post-fracture rehabilitation device according to claim 1, characterized in that, It also includes a filtering algorithm module; Both the mechanical sensor module and the healthy side data acquisition module are connected to the signal input terminal of the controller through the filtering algorithm module.

3. The post-fracture rehabilitation device according to claim 1, characterized in that, The joint module of the joint actuator is selected from at least one of the following: hip joint module, knee joint module, ankle joint module, shoulder joint module, wrist joint module, and elbow joint module; Each joint module has degrees of freedom matching the anatomical movement of that joint. The knee joint module supports flexion / extension with a single degree of freedom and an angle adjustment range of 0°-145°. The wrist joint module supports flexion / extension with a three-degree-of-freedom compound movement of 0°-90°, radial / ulnar deviation with a range of 0°-30°, and pronation / supination with a range of 0°-180°. The elbow joint module supports flexion / extension with a range of 0°-150° and pronation / supination with a range of 0°-180°. The ankle joint module supports dorsiflexion / plantar flexion with a range of 0°-45° and inversion / eversion with a range of 0°-20°.

4. The post-fracture rehabilitation device according to claim 1, characterized in that, The mechanical sensor module also includes a surface electromyography (EMG) sensor, which is connected to the signal input terminal of the controller. The controller includes a compensatory action recognition circuit for recognizing compensatory actions based on signals transmitted by the surface electromyography sensor.

5. A control method for a post-fracture rehabilitation device, characterized in that, The method, applied to the post-fracture rehabilitation device according to any one of claims 1 to 4, comprises the following steps: Step S1: Collect the mechanical parameters of the patient's healthy limb through the healthy side data acquisition module. The mechanical parameters include the isokinetic contraction force, isotonic contraction force, maximum contraction force and maximum range of motion of the healthy limb. The mechanical parameters are then transmitted to the controller as an individualized rehabilitation benchmark for the affected side training. Step S2: Set the initial driving parameters for training on the affected side using the controller based on the individualized rehabilitation benchmark. The initial driving parameters include the initial range of motion angles, the initial magnitude of assistive force or resistance, and a safety threshold. Step S3: Activate the joint dynamometer, and the drive component drives the affected limb to perform rehabilitation training movements. At the same time, the six-axis force sensor, angle sensor and acceleration sensor in the mechanical sensor module collect the resistance force, joint angle and movement speed data of the affected limb in real time and transmit them to the controller. Step S4: The controller compares the real-time collected biomechanical data of the affected limb with the individualized rehabilitation benchmark and safety threshold, and dynamically adjusts the driving parameters of the drive component according to the comparison results, including adjusting at least one of the following: range of motion angle, magnitude of assist force or resistance, and movement speed. Step S5: Repeat steps S3 to S4 until the preset training objective is achieved.

6. The control method for the post-fracture rehabilitation device according to claim 5, characterized in that, The method also includes steps for identifying and correcting compensatory actions: During training, at least one of the motion data and electromyographic signals of the affected limb is collected in real time through the mechanical sensor module. The controller determines whether there is a compensatory action based on the collected data. The compensatory action includes at least one of trajectory abnormality and compensatory force exertion. When a compensatory action is detected, the controller outputs a correction guidance signal.

7. The control method for the post-fracture rehabilitation device according to claim 6, characterized in that, The steps for determining whether a compensatory action exists include: The movement trajectory of the affected limb is collected by angle and accelerometer sensors and compared with the standard trajectory of the healthy side obtained in advance by the healthy side data acquisition module. When the trajectory deviation exceeds the preset angle, it is judged as trajectory abnormality.

8. The control method for the post-fracture rehabilitation device according to claim 6, characterized in that, The steps for determining whether a compensatory action exists include: Electromyography (EMG) signals of the target muscle group and synergistic muscle group in the affected limb are collected by a surface EMG sensor. When the EMG amplitude of the synergistic muscle group or non-target muscle group exceeds the preset ratio of the EMG amplitude of the target muscle group, it is determined to be compensatory exertion.

9. The control method for the post-fracture rehabilitation device according to claim 5, characterized in that, The method also includes fatigue monitoring and training adjustment steps: During training, the controller analyzes the peak force values ​​collected by the six-axis force sensor in real time and calculates the attenuation rate of the peak force values. When the attenuation rate exceeds a preset first fatigue threshold, the controller prompts the patient via voice or screen to increase rest time or shorten the duration of a single training session. When the attenuation rate exceeds a preset second fatigue threshold, the controller automatically reduces the output resistance or auxiliary force of the drive component, and / or forcibly terminates the current training and generates fatigue recovery suggestions.

10. A post-fracture rehabilitation system, characterized in that, include: Cloud server, patient terminal device, doctor terminal device, and post-fracture rehabilitation device as described in any one of claims 1 to 4; The post-fracture rehabilitation device further includes: a data transmission module, comprising a wireless communication chip and an antenna, wherein the wireless communication chip is electrically connected to the data output terminal of the controller; The patient terminal device is equipped with a patient-side application; The doctor's terminal device is equipped with a doctor's application; The cloud server is communicatively connected to both the patient terminal device and the doctor terminal device, and the data transmission module is communicatively connected to the cloud server via a wireless communication protocol.

Images