Intelligent rehabilitation device for scoliosis
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- THE SECOND AFFILIATED HOSPITAL OF HUNAN UNIV OF TRADITIONAL CHINESE MEDICINE
- Filing Date
- 2026-02-03
- Publication Date
- 2026-06-09
Smart Images
Figure CN122163381A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of scoliosis rehabilitation equipment technology, specifically to an intelligent scoliosis rehabilitation device. Background Technology
[0002] Scoliosis is a common three-dimensional structural deformity of the spine, characterized by asymmetrical changes in its morphology, usually accompanied by structural and biomechanical changes around the vertebral bodies. Currently, scoliosis is mainly diagnosed by measuring the Cobb angle on X-ray, with a Cobb angle >10° as the diagnostic criterion. Based on etiology, scoliosis can be classified into neuromuscular scoliosis, idiopathic scoliosis (AIS), congenital scoliosis, and scoliosis caused by other reasons; among them, idiopathic scoliosis is the most common, especially among adolescents.
[0003] Adolescents' bones are not yet fully mature and have high plasticity. Timely and scientific intervention plans after a diagnosis of scoliosis play an important role in improving deformity, stabilizing the spine, and relieving pain. They can even improve cardiopulmonary function and socio-psychological factors. Current treatment options for adolescent idiopathic scoliosis include static adjustment with a scoliosis brace and rehabilitation training using Schroth exercises and the Eight-Section Brocade exercise from traditional Chinese medicine to correct scoliosis. During training, patients need to remove the scoliosis brace and wear a rehabilitation training vest, such as the ScolioGold ActiveTrain Schroth training vest. This training vest has no rigid shell and uses elastic mesh fabric and zoned elastic traction straps to fit the patient's skin without restricting the midline of the spine. At the same time, it provides space for chest expansion to accommodate various postural changes during Schroth exercises or the Eight-Section Brocade exercise.
[0004] However, during actual rehabilitation training using the ScolioGold ActiveTrain vest, adolescent patients often perform self-rehabilitation exercises based on the rehabilitation movements prescribed by their doctors. Given the long treatment period, these self-directed exercises are prone to non-standard movements and insufficient completion. Furthermore, they are unable to determine whether their rehabilitation movements have reached the required standard level, leading to a lack of assessment by both patients and doctors regarding whether the target muscle groups have been adequately activated, thus reducing the effectiveness of the rehabilitation training.
[0005] Therefore, this invention proposes an intelligent rehabilitation device for scoliosis to solve the above problems. Summary of the Invention
[0006] To address the aforementioned issues, this invention provides an intelligent rehabilitation device for scoliosis. By real-time monitoring and analysis of electromyographic and pressure signals from the core muscle groups and key stress points of scoliosis, abnormal compensation is identified, corresponding rehabilitation plans are generated, and dynamic optimization is performed. This solves the shortcomings of traditional rehabilitation training, such as the difficulty in judging movement standards and monitoring muscle activation status.
[0007] To achieve the above objectives, the technical solution of the present invention is as follows: A smart rehabilitation device for scoliosis includes a vest body for rehabilitation training and an external operating platform. The vest body is configured with a double-layer structure, consisting of an inner vest layer and an outer vest layer. The inner vest layer is provided with a plurality of electromyography (EMG) receiving grooves and pressure receiving grooves. Fixing plates are slidably fitted into both the EMG receiving grooves and the pressure receiving grooves. A plurality of springs are fixedly connected to the side of the fixing plate near the outer vest layer. The other end of each spring is fixedly connected to the bottom wall of the corresponding EMG receiving groove and the pressure receiving groove. Limiting grooves are symmetrically provided on the inner sidewalls of both the EMG receiving grooves and the pressure receiving grooves. Sliders corresponding to the limiting grooves and located in the limiting grooves are fixedly connected to both sides of the fixing plate. SEMG EMG sensors are detachably connected to the fixing plates in the EMG receiving grooves, and pressure sensors are detachably connected to the fixing plates in the pressure receiving grooves.
[0008] The technical principle of the above solution is as follows: When the patient wears the vest, the body surface exerts pressure on the electromyography sensor and the pressure sensor. The pressure is transmitted to the corresponding fixing plate, and the fixing plate compresses the spring to form a buffer displacement, ensuring that the electromyography sensor and the pressure sensor are in close contact with the skin while avoiding excessive pressure. This allows the pressure sensor to directly collect contact pressure data, and the electromyography sensor to directly collect muscle electrical activity signals through the skin surface. The pressure signal and electromyography signal are then transmitted to the external operating table for processing.
[0009] The above approach has the following beneficial effects:
[0010] 1. This solution uses a spring to buffer the pressure of the sensor on the skin, and at the same time, the sensor is limited by the receiving groove and the spring force is applied to ensure that the sensor is always in contact with the patient's skin surface, thus ensuring the accuracy of pressure information and electromyographic information acquisition.
[0011] 2. This plan uses accurate electromyography and stress data to show the correspondence between muscle activation level and body stress distribution, providing a precise data basis for the formulation and adjustment of rehabilitation plans;
[0012] 3. This program can use electromyography (EMG) signals to provide feedback on whether the patient's training is adequate. For example, if there is a risk of compensation in the muscles where the EMG signal is abnormal, it indicates that the training movements are not performed correctly.
[0013] Furthermore, the electromyography receiving slots are respectively set on the inner layer of the vest at the positions of the erector spinae, multifidus, rectus abdominis and external oblique muscles on both sides of the human body.
[0014] Beneficial effects: It enables precise monitoring of the core muscle groups related to scoliosis and accurate identification of muscle compensation patterns.
[0015] Furthermore, pressure-receiving grooves are respectively set on the inner layer of the vest at the positions of the ribs, anterior superior iliac spine of the pelvis, armpits, waist, iliac crest of the pelvis, pectoral muscles, and lumbar muscles on both sides of the human body.
[0016] Beneficial effects: The pressure sensor's monitoring points cover key stress points in scoliosis, comprehensively monitoring the asymmetrical distribution of body pressure caused by scoliosis, and providing quantitative data for morphological assessment.
[0017] Furthermore, the control panel is equipped with a controller, which is connected to both the SEMG electromyography sensor and the pressure sensor.
[0018] Beneficial effects: The console provides doctors with an operating terminal for receiving and processing signal data from electromyography and pressure sensors in real time.
[0019] Furthermore, a touch display is embedded in the control panel, and the touch display is electrically connected to the controller.
[0020] Beneficial effects: Real-time data of electromyographic and pressure signals are displayed on a touch screen, and multi-dimensional biomechanical data are presented through a visual interface, helping doctors to intuitively and comprehensively understand the patient's scoliosis status and rehabilitation training progress.
[0021] Furthermore, the controller is equipped with a data processing module, a data analysis module, and a rehabilitation plan module;
[0022] The data processing module is used to process the signal data from the electromyography sensor and the pressure sensor;
[0023] The data analysis module is used to analyze the data results from the data processing module, analyze the correlation between electromyographic signals and pressure data, and assist doctors in judging the severity of the condition and the progress of rehabilitation.
[0024] The rehabilitation program module, based on the assessment report generated by the data analysis module, formulates corresponding intervention plans according to the individual differences of patients.
[0025] Furthermore, the data processing module includes a signal filtering unit and a data alignment unit.
[0026] The signal filtering and processing unit is used to filter the SEMG signal and remove noise interference;
[0027] The data alignment unit is used to precisely align the SEMG signal with the pressure distribution data in time, laying the foundation for data analysis.
[0028] Furthermore, the data analysis module includes a learning model unit and a dynamic evaluation unit.
[0029] The learning model unit establishes an electromyography-mechanical coupling analysis model through an LSTM-CNN hybrid neural network. The LSTM layer is used to process the temporal features of electromyography signals and extract muscle activation data. The CNN layer is used to analyze the spatial distribution features of the pressure sensor array and output the ratio of muscle activation on the concave and convex sides and the trunk pressure distribution asymmetry index to identify abnormal compensation situations.
[0030] The dynamic assessment unit is used to quantify the activation sequence, intensity symmetry, and pressure load distribution of the muscle groups on both sides of the spine, and generate an assessment report on the biomechanical characteristics of scoliosis, providing data support for doctors' diagnosis.
[0031] Furthermore, the rehabilitation program module includes a unit for combining traditional Chinese and Western medicine approaches and a program optimization unit.
[0032] The integrated Chinese and Western medicine approach unit is used to develop standardized intervention programs based on Schroeder exercise therapy and traditional Chinese medicine Baduanjin, and to generate personalized rehabilitation training plans by combining the standardized intervention programs with individual patient assessment reports.
[0033] The program optimization unit is used to dynamically adjust the training parameters of the personalized rehabilitation training plan by combining the personalized rehabilitation training plan with real-time feedback from the patient's training, so as to adapt to the training needs at different stages.
[0034] Beneficial Effects: Through the collaborative operation of the data processing module, data analysis module, and rehabilitation program module, the precision and personalization of scoliosis rehabilitation training were achieved. The data processing module effectively eliminated noise interference and ensured the temporal consistency of multi-source data through filtering and time alignment techniques, providing a high-quality data foundation for subsequent analysis. The data analysis module, with the help of the LSTM-CNN hybrid neural network model, deeply explored the temporal characteristics of electromyographic signals and the spatial distribution patterns of pressure data, accurately identified abnormal compensation patterns, and quantitatively assessed the biomechanical state of the spine. Based on the quantitative assessment results, the rehabilitation program module integrated Schroeder exercise therapy and traditional Chinese medicine Baduanjin (Eight Pieces of Brocade) to form a personalized training plan, and dynamically optimized training parameters through reinforcement learning algorithms, realizing a closed-loop intelligent management from data collection to intervention adjustment, significantly improving the scientific nature and effectiveness of rehabilitation training.
[0035] Furthermore, the vest body is fixedly connected with a fastening strap, and the fastening strap is equipped with Velcro.
[0036] Beneficial effects: The vest can be quickly fixed and its tightness adjusted by Velcro, which improves the ease of wearing and adapts to patients of different body types, ensuring that the sensor can maintain good contact with the patient's skin. Attached Figure Description
[0037] Figure 1 This is an axonometric view of the vest body of an embodiment of the intelligent rehabilitation device for scoliosis of the present invention;
[0038] Figure 2 This is a lateral sectional view of the electromyography cavity in an embodiment of the intelligent rehabilitation device for scoliosis of the present invention.
[0039] Figure 3 This is a schematic diagram showing the location of the front pressure sensor on the vest in an embodiment of the intelligent rehabilitation device for scoliosis of the present invention.
[0040] Figure 4 This is a schematic diagram showing the location of the pressure sensor on the back of the vest in an embodiment of the intelligent rehabilitation device for scoliosis of the present invention.
[0041] Figure 5 This is a schematic diagram showing the location of the pressure sensor on the left side of the vest in an embodiment of the intelligent rehabilitation device for scoliosis of the present invention.
[0042] Figure 6 This is a schematic diagram showing the location of the pressure sensor on the right side of the vest in an embodiment of the intelligent rehabilitation device for scoliosis of the present invention.
[0043] Figure 7 This is a schematic diagram of the controller logic of an embodiment of the intelligent rehabilitation device for scoliosis of the present invention.
[0044] The reference numerals in the accompanying drawings of the instruction manual include: 1. Vest body; 2. Inner layer of vest; 3. Outer layer of vest; 4. Electromyography reservoir; 5. Fixing plate; 6. Spring; 7. Limiting groove; 8. Slider; 9. Fixing strap; 10. Velcro. Detailed Implementation
[0045] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0046] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.
[0047] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.
[0048] The following detailed description illustrates the specific implementation method:
[0049] Example 1:
[0050] As attached Figure 1 As shown: A smart rehabilitation device for scoliosis includes a vest body 1 for wearing during rehabilitation training and an external operating table. The operating table is equipped with a controller, preferably an Advantech UNO-2484G. After the doctor guides the patient through rehabilitation exercises, the patient performs the rehabilitation training independently. Due to the young age of adolescent patients, the rehabilitation training is prone to non-standard movements and insufficient completion, making it impossible for the doctor to judge whether the patient's rehabilitation movements have met the standards. Consequently, it is impossible to determine whether the patient's target muscle groups have been activated, thus reducing the rehabilitation effect.
[0051] In order to enable real-time monitoring of patients' training parameters and provide a basis for doctors to judge rehabilitation effects and optimize subsequent rehabilitation plans, as shown in the appendix... Figure 2 As shown, the vest body 1 is designed with a double-layer structure, consisting of an inner vest layer 2 and an outer vest layer 3. The inner vest layer 2 has several electromyography (EMG) reservoirs 4 and pressure reservoirs. The EMG reservoirs 4 are respectively positioned on the inner vest layer 2 at locations corresponding to the erector spinae, multifidus, rectus abdominis, and external oblique muscles on both sides of the human body, as shown in the attached diagram. Figure 3 Appendix Figure 4 Appendix Figure 5 and attached Figure 6As shown, pressure receiving grooves are respectively set on the inner layer 2 of the vest at the positions of the ribs, anterior superior iliac spine of the pelvis, armpit, waist, iliac crest of the pelvis, pectoral muscles, and lumbar muscles on both sides of the human body; both the electromyography receiving groove 4 and the pressure receiving groove are slidably fitted with fixing plates 5. Several springs 6 are fixedly connected to the side of the fixing plates 5 near the outer layer 3 of the vest. The other end of each spring 6 is fixedly connected to the bottom wall of the corresponding electromyography receiving groove 4 and the pressure receiving groove. The inner sidewalls of both the electromyography receiving groove 4 and the pressure receiving groove are symmetrically provided with limiting grooves 7. Sliders 8 corresponding to the limiting grooves 7 are fixedly connected to both sides of the fixing plates 5. SEMG electromyography sensors can be detachably connected to the fixing plates 5 in the electromyography receiving groove 4, and pressure sensors can be detachably connected to the fixing plates 5 in the pressure receiving groove. Both the SEMG electromyography sensors and the pressure sensors are connected to the controller signal; a touch display is embedded in the operating table and is electrically connected to the controller; the preferred model of the SEMG electromyography sensor is the Delsys Trigno Avanti sensor, and the preferred model of the pressure sensor is the Interlink Electronics FSR. The preferred model for the 400 flexible pressure sensor and touch display is Siemens SIMATIC HMI KTP700 Basic.
[0052] As attached Figure 7 As shown, the controller includes a data processing module, a data analysis module, and a rehabilitation program module.
[0053] The data processing module is used to process the signal data from the electromyography sensor and the pressure sensor;
[0054] The data processing module includes a signal filtering unit and a data alignment unit.
[0055] The signal filtering and processing unit is used to filter the SEMG signal and remove noise interference. It uses a Butterworth bandpass filter (passband 20-500Hz) to remove power frequency interference (50Hz) and motion artifacts. At the same time, it uses an adaptive noise cancellation algorithm (based on the minimum mean square error criterion) to further suppress environmental noise and baseline drift in the electromyography signal, ensuring that the signal-to-noise ratio is ≥30dB.
[0056] The data alignment unit is used to accurately align the SEMG signal with the pressure distribution data in time, laying the foundation for data analysis. The Dynamic Time Warping (DTW) algorithm is used to calibrate the asynchronously acquired signals. A time mapping model is established based on the physiological correlation between muscle activation timing and pressure changes (such as the appearance of pressure peak after muscle contraction 0.2-0.5s), ultimately achieving a data time synchronization error of <1ms.
[0057] The data analysis module is used to analyze the data results from the data processing module, analyze the correlation between electromyographic signals and pressure data, and assist doctors in judging the severity of the condition and the progress of rehabilitation.
[0058] The data analysis module includes a learning model unit and a dynamic evaluation unit.
[0059] The learning model unit establishes an electromyography-mechanical coupling analysis model using an LSTM-CNN hybrid neural network. The LSTM layer processes the temporal features of the electromyography signals and extracts muscle activation data. The CNN layer analyzes the spatial distribution features of the pressure sensor array and outputs the ratio of muscle activation on the concave and convex sides and the trunk pressure distribution asymmetry index to identify abnormal compensation. The LSTM layer adopts a bidirectional structure (input dimension: 8-channel electromyography signal × 100ms sliding window) and captures the temporal dependence of muscle activation (such as the latency and duration of erector spinae contraction) through gating units. The CNN layer uses 3×3 convolutional kernels (input dimension: 14-channel pressure data × spatial grid) to extract the gradient features and asymmetry patterns of pressure distribution. The model training adopts a transfer learning strategy. After pre-training based on 500+ clinical data, it is fine-tuned for new patients in 10-20 iterations. The output results include: the RMS ratio of erector spinae on the concave and convex sides (normal reference value 0.9-1.1), the trunk pressure asymmetry index (normal reference value <5%), and the abnormal compensation probability score (such as the risk of overactivation of the left multifidus muscle).
[0060] The dynamic assessment unit quantifies the activation timing, intensity symmetry, and pressure load distribution of muscle groups on both sides of the spine, generating an assessment report on the biomechanical characteristics of scoliosis to provide data support for doctors' diagnosis. Based on the feature parameters output by the learning model, an assessment system is constructed: muscle activation symmetry indicators (e.g., contraction time difference of the left and right rectus abdominis muscles <100ms), pressure distribution balance indicators (e.g., pressure difference between the two sides of the pelvis <10kPa), and compensation pattern recognition indicators (e.g., premature activation of muscles on the convex side >200ms). The report uses a visual heatmap to display the electromyographic-pressure coupling relationship and uses red, yellow, and green to indicate the risk level (green: normal, yellow: mild abnormality, red: severe abnormality).
[0061] The rehabilitation program module, based on the assessment report generated by the data analysis module, formulates corresponding intervention plans according to the individual differences of patients.
[0062] The rehabilitation program module includes a unit combining traditional Chinese and Western medicine approaches and a program optimization unit.
[0063] The integrated TCM and Western medicine unit is used to develop standardized intervention plans based on Schroeder exercise therapy and Baduanjin (Eight Pieces of Brocade) from traditional Chinese medicine. It then combines these standardized intervention plans with individual patient assessment reports to generate personalized rehabilitation training plans. A movement library containing 24 Schroeder movements (such as quadrupedal posture and wall bar training) and 8 forms of Baduanjin is established. Each movement is labeled with biomechanical parameters (such as muscle activation threshold and target pressure distribution value). Targeted movements are matched based on abnormal indicators in the assessment report: for asymmetrical muscle activation, unilateral strengthening exercises such as single-knee kneeling are recommended; for imbalanced pressure distribution, spinal decompression movements such as lateral flexion sitting are added. The training plan generation uses a weighted scoring algorithm, allocating rest time (30-90 seconds) between movement sets and daily training frequency (2-4 sets) based on the degree of abnormality of the indicators (e.g., a Cobb angle of 30° corresponds to a weight of 0.7).
[0064] The program optimization unit is used to dynamically adjust the training parameters of the personalized rehabilitation training plan by combining the personalized rehabilitation training plan with real-time feedback from the patient's training, adapting to the training needs of different stages. It uses a reinforcement learning algorithm (DQN) to optimize the training parameters in real time, with "muscle symmetry improvement rate" and "pressure balance improvement rate" as reward functions. When a certain indicator fails to reach the preset threshold in three consecutive training sessions, it automatically adjusts the difficulty of the movement (such as increasing the squat depth of the wall bar training) or replaces the alternative movement (such as replacing the quadrupedal posture with the bird-dog posture). At the same time, it establishes a training effect prediction model, predicts the improvement trend of the Cobb angle in the next four weeks based on LSTM time series analysis, and adjusts the intervention intensity in advance (such as increasing the training time by 10% per week).
[0065] The specific implementation process is as follows: First, the electromyography (EMG) sensor and pressure sensor are respectively attached to the corresponding fixing plate 5, so that the probes of the EMG sensor and pressure sensor are located on the side closest to the patient's skin. During the rehabilitation training of adolescents, the patient first wears the vest body 1 used for rehabilitation training. The inner layer 2 of the vest is precisely set with several EMG receiving slots 4 and pressure receiving slots according to the core needs of human scoliosis rehabilitation. Among them, the EMG receiving slots 4 are set to correspond to the positions of the erector spinae, multifidus, rectus abdominis and external oblique muscles on both sides of the human body. These muscle groups are the key core muscle groups that maintain spinal stability and affect the effect of scoliosis correction. Monitoring allows for precise understanding of muscle activation status; the pressure-receiving grooves are positioned corresponding to the ribs on both sides of the body, the anterior superior iliac spine of the pelvis, the armpits, the waist, the iliac crest of the pelvis, the pectoral muscles, and the lumbar muscles. These areas are key regions for changes in body pressure distribution during scoliosis, and monitoring the pressure here can reflect the imbalance in biomechanical distribution caused by abnormal spinal morphology; considering that everyone's body shape is different, and the individual differences among adolescents are significant, in order to ensure that the sensor can accurately correspond to the target position, the vest body 1 is available in several sizes according to the adolescent's body shape. Before wearing, the doctor selects the vest body 1 that corresponds to the patient's body size for the patient to use;
[0066] During the process of the patient wearing the vest body 1, the skin will come into contact with the sensors in the electromyography (EMG) receiving groove 4 and the pressure receiving groove. At this time, the pressure generated by the skin on the EMG sensor and the pressure sensor will be transmitted to the corresponding fixing plate 5. Under the action of pressure, the fixing plate 5 will move towards the outer layer 3 of the vest, and at the same time compress the spring 6 fixed between the fixing plate 5 and the bottom wall of the receiving groove. The spring 6 plays a buffering role to avoid excessive pressure on the skin by the sensor and thus avoid discomfort, improving the patient's wearing comfort. On the other hand, the elasticity of the spring 6 can always push the fixing plate 5 and the EMG sensor and pressure sensor towards the skin, ensuring that the EMG sensor and pressure sensor are in close contact with the skin and ensuring the stability of data acquisition. At the same time, the limiting groove 7 on the inner side wall of the EMG receiving groove 4 and the pressure receiving groove slides with the slider 8 on the side wall of the fixing plate 5, which plays a limiting and guiding role in the movement of the fixing plate 5, preventing the fixing plate 5 from shifting during the movement, further ensuring the accuracy of the contact position between the EMG sensor and the pressure sensor and the skin, and avoiding data acquisition errors caused by positional shift.
[0067] When patients begin rehabilitation training as instructed by their doctors, the SEMG (electromyography) sensors within the electromyography (EMG) reservoir 4 collect EMG signals from the corresponding muscle groups in real time. These EMG signals directly reflect the activation level, contraction intensity, and contraction timing of the muscles. For example, when a patient performs a specific rehabilitation movement, the EMG signals of the target muscle group will exhibit corresponding characteristic changes. If the movement is not performed correctly, there may be insufficient EMG signal intensity in the target muscle group or abnormal activation of non-target muscle groups. Meanwhile, the pressure sensors within the pressure reservoir collect pressure data from the corresponding areas in real time. This data reflects the pressure distribution in different parts of the body. Due to spinal morphology abnormalities, scoliosis patients often experience asymmetrical pressure distribution in their bodies. By monitoring pressure data, the degree of this asymmetry can be quantified, and the effectiveness of rehabilitation training in correcting spinal morphology can be assessed. The controller's built-in data analysis module performs multi-dimensional analysis of the pressure data: First, it calculates the pressure difference between corresponding parts on both sides of the body. When the difference exceeds a preset threshold (e.g., 10 kPa), it is determined that the pressure distribution is asymmetrical. Second, it analyzes the offset of the pressure center through a pressure distribution heatmap. If the pressure center deviates from the midline of the spine by more than 5 cm, it indicates that the mechanical imbalance caused by scoliosis is relatively serious. At the same time, it combines time series analysis to monitor the dynamic trend of pressure data. If the degree of pressure asymmetry does not show a gradual decreasing trend during training, it is determined that the current rehabilitation training is not effective in correcting spinal morphology.
[0068] The signals collected by the SEMG electromyography (EMG) and pressure sensors are transmitted in real time to the controller in the external operating console. The controller, as the core processing unit of the entire device, receives and performs preliminary processing of these signals, and then displays the processed EMG signals and pressure data in real time on a touchscreen display. The specific processing and analysis process is as follows: First, the signal filtering unit uses a Butterworth bandpass filter (passband 20-500Hz) to remove power frequency interference (50Hz) and motion artifacts. Then, an adaptive noise cancellation algorithm based on the minimum mean square error criterion further suppresses environmental noise and baseline drift in the EMG signals. To ensure a stable signal-to-noise ratio (SNR) of ≥30dB for electromyography (EMG) signals, a dynamic time warping (DTW) algorithm is used via a data alignment unit to establish a time mapping model based on the physiological correlation between the peak pressure occurring 0.2-0.5s after muscle contraction, achieving a time synchronization error of <1ms between EMG signals and pressure data. Finally, a deep analysis is performed using an LSTM-CNN hybrid neural network model. The bidirectional LSTM layer captures the temporal dependence features of the EMG signals, while the CNN layer extracts the spatial distribution features of the pressure data, outputting key parameters such as muscle activation level and pressure distribution asymmetry index. Doctors can visually observe the patient's training process via a touchscreen display. Dynamic changes in muscle activation and body pressure distribution are used to determine whether a patient's rehabilitation movements are standard and whether the target muscle groups are properly activated. For example, if the electromyographic signal intensity of a target muscle group is consistently below the normal range, it suggests insufficient activation of that muscle group, potentially indicating inadequate movement completion. Similarly, a significant difference in pressure data between corresponding areas on both sides of the body suggests that the biomechanical imbalance caused by scoliosis has not been effectively improved. For instance, if the RMS value of an electromyographic signal for a target muscle group is consistently below 80% of the normal range, it indicates insufficient activation of that muscle group, potentially indicating inadequate movement completion. When the pressure difference between corresponding parts of the body on both sides exceeds 15 kPa for 10 consecutive seconds, it indicates that the biomechanical imbalance caused by scoliosis has not been effectively improved. When abnormally elevated electromyographic signals of non-target muscle groups (such as the trapezius) are detected, exceeding 120% of the normal range, it suggests that the patient may have abnormal compensatory movements that need to be corrected in time. By monitoring the patient's training parameters in real time and accurately, we can provide doctors with objective and accurate evidence to judge the rehabilitation effect, and at the same time provide data support for the optimization and adjustment of subsequent rehabilitation plans. This will help improve the rehabilitation training effect of adolescent scoliosis and help patients better improve their scoliosis condition.
[0069] Example 2:
[0070] As attached Figure 1As shown, the difference from Embodiment 1 is that a fixing strap 9 is fixedly connected to the vest body 1, and the fixing strap 9 is provided with Velcro 10; the Velcro 10 enables the vest to be quickly fixed and the tightness adjusted to fit patients of different body types. The adjustment of the tightness of the fixing strap 9 ensures that the electromyography sensor and pressure sensor can stay in contact with the patient's skin, thereby improving the accuracy of electromyography and pressure data acquisition.
[0071] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.
Claims
1. A scoliosis intelligent rehabilitation device, comprising a vest body (1) for wearing during rehabilitation training and an external operating table, characterized in that: The vest body (1) is set as a double-layer structure. The double-layer structure of the vest body (1) is the inner layer (2) and the outer layer (3). The inner layer (2) of the vest is provided with several electromyography (EMG) receiving grooves (4) and pressure receiving grooves. Fixing plates (5) are slidably fitted in the EMG receiving grooves (4) and the pressure receiving grooves. Several springs (6) are fixedly connected to the side of the fixing plate (5) near the outer layer (3). The other end of the springs (6) is fixedly connected to the bottom wall of the corresponding EMG receiving groove (4) and the pressure receiving groove. Limiting grooves (7) are symmetrically provided on the inner sidewalls of the EMG receiving grooves (4) and the pressure receiving grooves. Sliders (8) corresponding to the limiting grooves (7) are fixedly connected to both sides of the fixing plate (5). SEMG EMG sensors can be detachably connected to the fixing plate (5) in the EMG receiving groove (4). Pressure sensors can be detachably connected to the fixing plate (5) in the pressure receiving groove.
2. The intelligent rehabilitation device for scoliosis according to claim 1, characterized in that: The electromyography reservoirs (4) are respectively set on the inner layer (2) of the vest, corresponding to the erector spinae, multifidus, rectus abdominis and external oblique muscles on both sides of the human body.
3. The intelligent rehabilitation device for scoliosis according to claim 1, characterized in that: Pressure-receiving grooves are respectively set on the inner layer of the vest (2) at the positions of the ribs, anterior superior iliac spine of the pelvis, armpit, waist, iliac crest of the pelvis, pectoral muscles and lumbar muscles on both sides of the human body.
4. The intelligent rehabilitation device for scoliosis according to claim 1, characterized in that: The control panel is equipped with a controller, which is connected to both the SEMG electromyography sensor and the pressure sensor.
5. The intelligent rehabilitation device for scoliosis according to claim 4, characterized in that: The control panel has an embedded touch screen display, which is electrically connected to the controller.
6. The intelligent rehabilitation device for scoliosis according to claim 5, characterized in that: The controller is equipped with a data processing module, a data analysis module, and a rehabilitation program module; The data processing module is used to process the signal data from the electromyography sensor and the pressure sensor; The data analysis module is used to analyze the data results from the data processing module, analyze the correlation between electromyographic signals and pressure data, and assist doctors in judging the severity of the condition and the progress of rehabilitation. The rehabilitation program module, based on the assessment report generated by the data analysis module, formulates corresponding intervention plans according to the individual differences of patients.
7. The intelligent rehabilitation device for scoliosis according to claim 6, characterized in that: The data processing module includes a signal filtering unit and a data alignment unit. The signal filtering and processing unit is used to filter the SEMG signal and remove noise interference; The data alignment unit is used to precisely align the SEMG signal with the pressure distribution data in time, laying the foundation for data analysis.
8. The intelligent rehabilitation device for scoliosis according to claim 6, characterized in that: The data analysis module includes a learning model unit and a dynamic evaluation unit. The learning model unit establishes an electromyography-mechanical coupling analysis model through an LSTM-CNN hybrid neural network. The LSTM layer is used to process the temporal features of electromyography signals and extract muscle activation data. The CNN layer is used to analyze the spatial distribution features of the pressure sensor array and output the ratio of muscle activation on the concave and convex sides and the trunk pressure distribution asymmetry index to identify abnormal compensation situations. The dynamic assessment unit is used to quantify the activation sequence, intensity symmetry, and pressure load distribution of the muscle groups on both sides of the spine, and generate an assessment report on the biomechanical characteristics of scoliosis, providing data support for doctors' diagnosis.
9. The intelligent rehabilitation device for scoliosis according to claim 6, characterized in that: The rehabilitation program module includes a unit combining traditional Chinese and Western medicine approaches and a program optimization unit. The integrated Chinese and Western medicine approach unit is used to develop standardized intervention programs based on Schroeder exercise therapy and traditional Chinese medicine Baduanjin, and to generate personalized rehabilitation training plans by combining the standardized intervention programs with individual patient assessment reports. The program optimization unit is used to dynamically adjust the training parameters of the personalized rehabilitation training plan by combining the personalized rehabilitation training plan with real-time feedback from the patient's training, so as to adapt to the training needs at different stages.
10. The intelligent rehabilitation device for scoliosis according to claim 1, characterized in that: A fixing strap (9) is fixedly connected to the vest body (1), and Velcro (10) is provided on the fixing strap (9).