Air bag control method, medium and system for orthopedic force of scoliosis orthosis
By using multiple sensors in scoliosis orthosis to acquire signals, establish distribution models and optimization algorithms, and dynamically adjust airbag pressure, the inefficiency problem in the existing technology that relies on empirical judgment is solved, and correction effect, comfort and power are achieved Optimization of consumption.
Patent Information
- Application Number
- CN202411118397.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-15
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-08-15
AI Technical Summary
The existing methods of airbag pressure regulation of scoliosis orthotics rely on the clinician's empirical judgment, lack scientific analysis, is inefficient, and is difficult to achieve precise control.
Multiple pressure sensors, vibration sensors and air pressure sensors are used to collect signals, establish pressure distribution, vibration and airbag pressure distribution models, combine attitude characteristics, and dynamically adjust the airbag pressure through a multi-objective optimization algorithm to achieve adaptive control of the orthotic.
It improves the correction effect and wear comfort, reduces power consumption, and realizes precise control and intelligent management of orthotic correction force.
Smart Images

Figure CN119015030A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of scoliosis orthoses, and in particular, relates to an airbag control method, medium and system for the orthopedic force of a scoliosis orthosis. Background Art
[0002] Scoliosis is a common spinal deformity disease, the main feature of which is the sideways curvature of the spine. According to statistics, the incidence of scoliosis is about 2%-3%, and it often occurs in adolescence, causing serious impact on the patient's physical and mental health and quality of life. Therefore, it is crucial to take effective corrective treatment in a timely manner.
[0003] At present, the commonly used methods for correcting scoliosis are mainly surgical treatment and conservative treatment. Surgical treatment methods include posterior correction and fusion, anterior and posterior correction and fusion, etc., which can effectively correct severe scoliosis deformity. However, the operation is traumatic, the recovery period is long, and there is a certain risk of surgical complications, so it is mostly used for patients with more serious conditions. In contrast, conservative treatment is more suitable for patients with milder conditions, mainly including orthotic correction, physical therapy, rehabilitation training, etc.
[0004] Among them, orthotic correction is the most common method of conservative treatment. Orthotics gradually correct scoliosis deformity by applying a certain corrective force and are widely used in clinical practice. For example, the utility model CN202210471088.X discloses a spin-type scoliosis orthosis, the scoliosis orthosis disclosed in CN202023005647.3, the intelligent scoliosis orthosis disclosed in CN202023150373.7, and the intelligent orthotic device disclosed in CN202020245828.4; common orthotics include Boston type, Milwaukee type, Cheneau type, etc., which are mainly made of rigid materials and can provide stable corrective force. In order to improve the correction effect and wearing comfort, the orthosis often contains an adjustable airbag, which regulates the size and distribution of the corrective force by adjusting the pressure in the airbag.
[0005] However, the existing orthosis airbag pressure adjustment methods mainly rely on the clinical physician's experience and judgment, lacking scientific analysis basis. Doctors usually manually adjust the airbag pressure based on the patient's subjective feedback, which is inefficient and difficult to achieve precise control. Summary of the invention
[0006] In view of this, the present invention provides an airbag control method, medium and system for the corrective force of a scoliosis orthosis, which can solve the technical problems that the pressure regulation of the existing method mainly relies on the experience and judgment of the clinician, lacks scientific analysis basis, is inefficient, and is difficult to achieve precise control.
[0007] The present invention is achieved in that:
[0008] A first aspect of the present invention provides an airbag control method for the corrective force of a scoliosis orthosis, comprising the following steps:
[0009] S10, acquiring pressure signals, vibration signals, and air pressure signals collected by multiple pressure sensors, multiple vibration sensors, and multiple air pressure sensors to form a pressure signal set, a vibration signal set, and an air pressure signal set;
[0010] S20, establishing a pressure distribution model of the contact surface between the orthosis and the skin according to the pressure signal set and the position of each pressure sensor;
[0011] S30, establishing a vibration model of the orthosis according to the vibration signal set and the position of each vibration sensor, so as to evaluate the stability and comfort of the orthosis;
[0012] S40, establishing an airbag pressure distribution model according to the air pressure signal set and the position of each airbag, so as to adjust the pressure in the airbag;
[0013] S50, obtaining pressure distribution characteristics, vibration characteristics and air pressure distribution characteristics of the pressure distribution model, vibration model and airbag pressure distribution model under different postures of the user, and establishing a posture-feature mapping relationship model for predicting the orthopedic effect under different postures;
[0014] S60, establishing a multi-objective optimization model, with the objectives of maximizing the correction effect, optimizing the wearing comfort and minimizing the power consumption, with the airbag pressure as the decision variable, and with the pressure distribution, vibration and air pressure safety threshold as the constraint conditions;
[0015] S70, solving the multi-objective optimization model to obtain the optimal airbag pressure configuration scheme under different postures;
[0016] S80, dynamically adjusting the pressure of each airbag according to the optimal airbag pressure configuration scheme and in combination with the user's posture detected in real time, so as to achieve adaptive control of the orthotic force of the orthosis.
[0017] Wherein, the step S10 specifically includes:
[0018] Step 101: Arrange a plurality of pressure sensors on the contact surface between the orthosis and the skin to collect pressure signals on the contact surface and record the position coordinate information of each pressure sensor;
[0019] Step 102: Arrange a plurality of vibration sensors on the outer surface of the orthosis to collect vibration signals of the orthosis and record the position coordinate information of each vibration sensor;
[0020] Step 103: Arrange an air pressure sensor at the vent of the airbag to collect the air pressure signal in the airbag;
[0021] Step 104: The collected pressure signals, vibration signals and air pressure signals are respectively formed into a pressure signal set, a vibration signal set and an air pressure signal set.
[0022] The specific steps of step S20 include:
[0023] Step 201: Using an interpolation algorithm, based on the pressure signals collected by each pressure sensor and their position coordinates, a pressure distribution model of the contact surface between the orthosis and the skin is constructed, and the model can describe the pressure value at any position on the contact surface;
[0024] Step 202: The pressure distribution model can be used to analyze the pressure characteristics of the orthosis applied to the patient's skin in different postures, providing a basis for optimizing the airbag pressure configuration.
[0025] Wherein, the specific steps of step S30 include:
[0026] Step 301: Using vibration analysis techniques such as modal analysis, a vibration model of the orthosis is established based on the vibration signals collected by each vibration sensor and their position coordinates. The model can describe the vibration characteristics of the orthosis at different positions, such as amplitude, frequency, etc.;
[0027] Step 302: The vibration model can be used to evaluate the stability and wearing comfort of the orthosis, for example, to identify high amplitude areas near the resonant frequency and adjust the airbag pressure configuration to suppress these vibrations.
[0028] The specific steps of step S40 include:
[0029] Step 401: Using an interpolation algorithm, an airbag pressure distribution model is constructed based on the air pressure signals collected by each air pressure sensor and its position coordinates. The model can describe the pressure value in each airbag.
[0030] Step 402: The air pressure distribution model provides a basis for adjusting the pressure in each airbag. It can identify areas with low or high pressure according to the pressure distribution characteristics and appropriately adjust the pressure of the corresponding airbag.
[0031] The specific steps of step S50 include:
[0032] Step 501: collect motion signals of the user in different postures through an acceleration sensor or a gyroscope arranged on the outer surface of the orthosis, and identify different posture features;
[0033] Step 502: Combine the pressure distribution model, vibration model and air pressure distribution model established in steps S20-S40 to extract the pressure distribution characteristics, vibration characteristics and air pressure distribution characteristics of each model in different postures;
[0034] Step 503: Use a supervised learning algorithm to establish a posture-feature mapping relationship model, which can predict the corresponding pressure distribution, vibration characteristics and air pressure distribution based on the user's posture detected in real time.
[0035] The specific steps of step S60 include:
[0036] Step 601: taking correction effect, wearing comfort and power consumption as optimization objective functions, wherein the correction effect can be characterized by a pressure distribution index, the wearing comfort can be characterized by a vibration characteristic index, and the power consumption can be characterized by an airbag pressure index;
[0037] Step 602: Use the pressure value of each airbag as a decision variable, and set safety threshold constraints for pressure distribution, vibration characteristics, and air pressure value, such as the maximum pressure does not exceed 40 kPa, the maximum vibration acceleration does not exceed 5 m / s^2, and the maximum airbag pressure does not exceed 80 kPa;
[0038] Step 603: Use a multi-objective optimization algorithm to solve the optimization model and obtain the optimal airbag pressure configuration solution under different postures.
[0039] The specific steps of step S70 include:
[0040] Step 701: using a multi-objective optimization algorithm, such as a genetic algorithm or a particle swarm optimization algorithm, to solve the multi-objective optimization model established in step S60;
[0041] Step 702: The goal of the algorithm is to maximize the correction effect, optimize the wearing comfort, and minimize the power consumption under the premise of satisfying the pressure distribution, vibration characteristics, and air pressure safety constraints;
[0042] Step 703: Obtain the optimal airbag pressure configuration scheme under different postures through iterative optimization.
[0043] The specific steps of step S80 include:
[0044] Step 801: monitor the posture changes of the user in real time, and use the posture-feature mapping relationship model established in step S50 to predict the corresponding pressure distribution, vibration characteristics and air pressure distribution;
[0045] Step 802: Compare the prediction result with the optimal airbag pressure configuration solution obtained in step S70 to determine the airbag that needs to be adjusted and its pressure value;
[0046] Step 803: adopt a progressive adjustment strategy to gradually adjust the pressure of each airbag to avoid discomfort caused by sudden pressure changes;
[0047] Step 804: Set up a safety monitoring mechanism. When excessive pressure or abnormal vibration is detected, immediately perform airbag decompression to ensure safe use.
[0048] Wherein, the multiple pressure sensors are multiple pressure sensors arranged on the contact surface between the orthosis and the skin.
[0049] Wherein, the multiple vibration sensors are multiple vibration sensors arranged on the outer surface of the orthosis.
[0050] Wherein, the air pressure sensor is arranged at the vent of the airbag to collect the air pressure in the airbag.
[0051] The user's posture is detected and acquired by an acceleration sensor or a gyroscope arranged on the outer surface of the orthosis.
[0052] Wherein, the posture-feature mapping relationship model adopts a multivariate linear regression model.
[0053] Wherein, the method used to solve the multi-objective optimization model is a genetic algorithm.
[0054] When a genetic algorithm is used to solve the multi-objective optimization model, the fitness function is a similarity function of a vector composed of each objective.
[0055] A second aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program instructions, and when the program instructions are executed, they are used to execute the above-mentioned airbag control method for the corrective force of a scoliosis orthosis.
[0056] A third aspect of the present invention provides an airbag control system for the corrective force of a scoliosis orthosis, which includes the above-mentioned computer-readable storage medium.
[0057] Compared with the prior art, the airbag control method, medium and system of the corrective force of the scoliosis orthosis provided by the present invention have the following beneficial effects: 1. Improved correction effect. The method establishes a pressure distribution model, monitors the pressure characteristics of the contact surface between the orthosis and the skin in real time, and adjusts the pressure of each airbag according to the pressure distribution characteristics, effectively improving the accuracy and correction effect of the correction force. Compared with the traditional adjustment method relying on the doctor's experience, this method can achieve more refined correction force control.
[0058] 2. Enhanced wearing comfort. This method not only takes into account the pressure distribution, but also establishes a vibration model for the orthosis to evaluate its vibration characteristics in real time. By optimizing the airbag pressure configuration, it can not only meet the requirements of the correction effect, but also suppress the local high vibration area, greatly improving the user's comfort.
[0059] 3. Reduced power consumption. This method adopts a multi-objective optimization approach and takes minimizing power consumption as one of the optimization goals. By intelligently adjusting the pressure of each airbag, power consumption is minimized and the use efficiency is significantly improved while meeting the correction effect and comfort requirements.
[0060] 4. Adaptive control is achieved. This method combines the user's real-time posture detection and establishes a posture-feature mapping relationship model. It can predict the corresponding pressure distribution, vibration characteristics and air pressure distribution according to different postures, and dynamically adjust the pressure of each airbag accordingly to achieve adaptive control of the correction force of the orthosis. This adaptive adjustment strategy greatly improves the intelligence level of the entire correction system.
[0061] In general, the adaptive control method proposed in the present invention makes full use of multidisciplinary technical means, including sensing technology, signal processing, optimization algorithms, etc., and comprehensively considers the correction effect, wearing comfort and power consumption, and solves the technical problems that the pressure regulation of the existing method mainly relies on the experience and judgment of the clinician, lacks scientific analysis basis, is inefficient, and is difficult to achieve precise control. BRIEF DESCRIPTION OF THE DRAWINGS
[0062] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.
[0063] Figure 1 A flow chart of the method provided by the present invention; DETAILED DESCRIPTION
[0064] In order to make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.
[0065] like Figure 1 FIG. 1 is a flow chart of an airbag control method for the orthopedic force of a scoliosis orthosis provided by the present invention. The method comprises the following steps:
[0066] S10, acquiring pressure signals, vibration signals, and air pressure signals collected by multiple pressure sensors, multiple vibration sensors, and multiple air pressure sensors to form a pressure signal set, a vibration signal set, and an air pressure signal set;
[0067] S20, establishing a pressure distribution model of the contact surface between the orthosis and the skin according to the pressure signal set and the position of each pressure sensor;
[0068] S30, establishing a vibration model of the orthosis according to the vibration signal set and the position of each vibration sensor, so as to evaluate the stability and comfort of the orthosis;
[0069] S40, establishing an airbag pressure distribution model according to the air pressure signal set and the position of each airbag, so as to adjust the pressure in the airbag;
[0070] S50, obtaining pressure distribution features, vibration features and air pressure distribution features of the pressure distribution model, the vibration model and the airbag pressure distribution model under different postures of the user, and establishing a posture-feature mapping relationship model for predicting the correction effect under different postures;
[0071] S60, establishing a multi-objective optimization model, with the objectives of maximizing the correction effect, optimizing the wearing comfort and minimizing the power consumption, with the airbag pressure as the decision variable, and with the pressure distribution, vibration and air pressure safety threshold as the constraint conditions;
[0072] S70, solving a multi-objective optimization model to obtain an optimal airbag pressure configuration solution under different postures;
[0073] S80. According to the optimal airbag pressure configuration scheme and in combination with the user's posture detected in real time, the pressure of each airbag is dynamically adjusted to achieve adaptive control of the orthotic force of the orthosis.
[0074] The specific implementation methods of the above steps are described in detail below:
[0075] Step S10: Obtain pressure signals, vibration signals and air pressure signals collected by multiple pressure sensors, multiple vibration sensors and multiple air pressure sensors to form a pressure signal set, a vibration signal set and an air pressure signal set.
[0076] The specific implementation is:
[0077] 1. Multiple pressure sensors are placed on the contact surface between the orthosis and the skin to collect pressure signals on the contact surface. The position coordinate information of each pressure sensor also needs to be recorded.
[0078] 2. Multiple vibration sensors are placed on the outer surface of the orthosis to collect the vibration signal of the orthosis. The position coordinate information of each vibration sensor also needs to be recorded.
[0079] 3. Install an air pressure sensor at the vent of the airbag to collect the air pressure signal inside the airbag.
[0080] 4. The collected pressure signals, vibration signals and air pressure signals are respectively formed into a pressure signal set, a vibration signal set and an air pressure signal set. These signal sets provide basic data for the subsequent establishment of pressure distribution models, vibration models and air pressure distribution models.
[0081] Step S20: Establishing a pressure distribution model of the contact surface between the orthosis and the skin according to the pressure signal set and the position of each pressure sensor.
[0082] The specific implementation is:
[0083] 1. Use interpolation algorithms, such as bilinear interpolation or adaptive spline interpolation, to construct a pressure distribution model of the contact surface between the orthosis and the skin based on the pressure signals collected by each pressure sensor and its position coordinates. This model can describe the pressure value at any position on the contact surface.
[0084] 2. The pressure distribution model can be used to analyze the pressure characteristics of the orthosis on the patient's skin in different postures, providing a basis for optimizing the airbag pressure configuration. For example, local high-pressure areas can be identified based on the pressure distribution characteristics to adjust the pressure of the corresponding airbag in a targeted manner.
[0085] Step S30: Establishing a vibration model of the orthosis based on the vibration signal set and the position of each vibration sensor, so as to evaluate the stability and comfort of the orthosis.
[0086] The specific implementation is:
[0087] 1. Using vibration analysis techniques such as modal analysis, a vibration model of the orthosis is established based on the vibration signals collected by each vibration sensor and its position coordinates. This model can describe the vibration characteristics of the orthosis at different positions, such as amplitude, frequency, etc.
[0088] 2. The vibration model can be used to evaluate the stability and wearing comfort of the orthosis. For example, the high amplitude area near the resonant frequency can be identified based on the vibration characteristics, and the airbag pressure configuration can be adjusted to suppress these vibrations, thereby improving the stability of the orthosis and the comfort of the user.
[0089] Step S40: Establish an airbag pressure distribution model based on the air pressure signal set and the position of each airbag to adjust the pressure in the airbag.
[0090] The specific implementation is:
[0091] 1. Use interpolation algorithms, such as bilinear interpolation or adaptive spline interpolation, to construct an airbag pressure distribution model based on the air pressure signals collected by each air pressure sensor and its position coordinates. This model can describe the pressure value in each airbag.
[0092] 2. The air pressure distribution model provides a basis for adjusting the pressure in each airbag. For example, the pressure distribution characteristics can be used to identify areas with low pressure and appropriately increase the pressure of the corresponding airbag, or to identify areas with excessively high pressure and appropriately reduce the pressure of the corresponding airbag to achieve the goal of overall pressure optimization.
[0093] Step S50: Obtain the pressure distribution characteristics, vibration characteristics and air pressure distribution characteristics of the pressure distribution model, vibration model and airbag pressure distribution model under different postures of the user, and establish a posture-feature mapping relationship model for predicting the orthopedic effect under different postures.
[0094] The specific implementation is:
[0095] 1. Through the acceleration sensor or gyroscope set on the outer surface of the orthosis, the user's motion signals in different postures are collected and different posture characteristics are identified.
[0096] 2. Combine the pressure distribution model, vibration model and air pressure distribution model established in steps S20-S40 to extract the pressure distribution characteristics, vibration characteristics and air pressure distribution characteristics of each model under different postures.
[0097] 3. Use supervised learning algorithms, such as multivariate linear regression or neural networks, to establish a posture-feature mapping relationship model. This model can predict the corresponding pressure distribution, vibration characteristics and air pressure distribution based on the user's posture detected in real time, thereby predicting the orthopedic effect under different postures.
[0098] Step S60: Establish a multi-objective optimization model with the goals of maximizing the correction effect, optimizing the wearing comfort and minimizing the power consumption, with the airbag pressure as the decision variable, and the pressure distribution, vibration and air pressure safety thresholds as constraints.
[0099] The specific implementation is:
[0100] 1. The correction effect, wearing comfort and power consumption are taken as optimization objective functions. The correction effect can be characterized by pressure distribution index, wearing comfort can be characterized by vibration characteristic index, and power consumption can be characterized by airbag pressure index.
[0101] 2. Use the pressure value of each airbag as a decision variable. The pressure distribution, vibration characteristics and air pressure value must meet certain safety threshold constraints, such as:
[0102] -The maximum value of the pressure distribution does not exceed 40kPa
[0103] -The maximum value of vibration acceleration does not exceed 5m / s^2
[0104] -The maximum airbag pressure does not exceed 80kPa
[0105] 3. Use a multi-objective optimization algorithm, such as a genetic algorithm or a particle swarm optimization algorithm, to solve this multi-objective optimization model and obtain the optimal airbag pressure configuration scheme under different postures. This scheme can take into account the optimization of correction effect, wearing comfort and power consumption.
[0106] Step S70: Solve the multi-objective optimization model to obtain the optimal airbag pressure configuration solution under different postures.
[0107] The specific implementation is:
[0108] 1. Use a multi-objective optimization algorithm, such as a genetic algorithm or a particle swarm optimization algorithm, to solve the multi-objective optimization model established in step S60.
[0109] 2. The goal of the algorithm is to maximize the correction effect, optimize wearing comfort, and minimize power consumption while satisfying the pressure distribution, vibration characteristics, and air pressure safety constraints.
[0110] 3. Through iterative optimization, the optimal airbag pressure configuration scheme under different postures can be obtained. This scheme provides the best reference for the subsequent dynamic adjustment of airbag pressure.
[0111] Step S80: According to the optimal airbag pressure configuration scheme, combined with the user's posture detected in real time, the pressure of each airbag is dynamically adjusted to achieve adaptive control of the orthotic force of the orthosis.
[0112] The specific implementation is:
[0113] 1. Monitor the user's posture changes in real time, and predict the corresponding pressure distribution, vibration characteristics and air pressure distribution through the posture-feature mapping relationship model established in step S50.
[0114] 2. Compare the prediction result with the optimal airbag pressure configuration plan obtained in step S70 to determine the airbag that needs to be adjusted and its pressure value.
[0115] 3. Use a progressive adjustment strategy to gradually adjust the pressure of each airbag to avoid discomfort caused by sudden changes in pressure.
[0116] 4. At the same time, a safety monitoring mechanism is set up. When excessive pressure or abnormal vibration is detected, the airbag will be decompressed immediately to ensure safe use.
[0117] 5. By dynamically adjusting the airbag pressure, adaptive control of the orthotic force of the orthosis is achieved to improve the correction effect, enhance wearing comfort, and reduce power consumption.
[0118] In summary, the patent claims describe an adaptive control method for scoliosis braces based on airbag pressure. A multi-objective optimization model was established by modeling and analyzing pressure, vibration, and air pressure signals, and dynamic airbag pressure regulation was achieved in combination with real-time posture detection, thereby improving the correction effect, wearing comfort, and power efficiency. This method integrates multidisciplinary knowledge, including sensor technology, signal processing, mechanical vibration, optimization algorithms, etc., and provides an innovative solution for the intelligent control of scoliosis braces.
[0119] A second aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program instructions, and when the program instructions are executed, they are used to execute the above-mentioned airbag control method for the corrective force of a scoliosis orthosis.
[0120] A third aspect of the present invention provides an airbag control system for the corrective force of a scoliosis orthosis, which includes the above-mentioned computer-readable storage medium.
[0121] In order to better understand and implement the present invention, the following provides a detailed description of a specific implementation of the present invention in combination with formulas. Of course, this detailed description can also be directly used in a computer program, as a specific implementation of the present invention in a computer-readable storage medium, or in an electronic device such as a computer, a single-chip microcomputer, or a control chip, and is specifically described as follows:
[0122] Step S10: Obtain pressure signals, vibration signals and air pressure signals collected by multiple pressure sensors, multiple vibration sensors and multiple air pressure sensors to form a pressure signal set, a vibration signal set and an air pressure signal set.
[0123] The specific implementation is as follows:
[0124] 1. Pressure signal acquisition: Assume that there are n pressure sensors on the contact surface between the orthosis and the skin, and the position coordinates of each sensor are (x i ,y i ), i = 1, 2, ..., n. The pressure signal collected by the i-th pressure sensor is recorded as P i (t), the pressure signal set can be expressed as:
[0125] P = {P1(t), P2(t), ..., P n (t)}
[0126] Among them, P i(t) represents the pressure value collected by the i-th pressure sensor at time t. The position coordinates (x i ,y i ) also needs to be recorded for subsequent establishment of the pressure distribution model.
[0127] 2. Vibration signal acquisition: Assume that there are m vibration sensors on the outer surface of the orthosis, and the position coordinates of each sensor are (x j ,y j ), j = 1, 2, ..., m. The vibration signal collected by the jth vibration sensor is denoted as V j (t), then the vibration signal set can be expressed as:
[0128] V={V1(t), V2(t), ..., V m (t)}
[0129] Among them, V j (t) represents the vibration value collected by the jth vibration sensor at time t. The position coordinates of each vibration sensor (x j ,y j ) also need to be recorded for subsequent establishment of the vibration model.
[0130] 3. Air pressure signal acquisition: Assume that there are k airbags on the orthosis, and each airbag has an air pressure sensor at its vent. The air pressure signal collected by the lth air pressure sensor is denoted as G l (t), the air pressure signal set can be expressed as:
[0131] G={G1(t), G2(t), …, G k (t)}
[0132] Among them, G l (t) represents the air pressure value collected by the lth air pressure sensor at time t. The location information of each air pressure sensor also needs to be recorded for subsequent establishment of the air pressure distribution model.
[0133] Through the above steps, the pressure signal set P, the vibration signal set V and the air pressure signal set G can be obtained, providing basic data for the subsequent establishment of various distribution models.
[0134] Step S20: establishing a pressure distribution model of the contact surface between the orthosis and the skin according to the pressure signal set and the position of each pressure sensor.
[0135] The specific implementation is as follows:
[0136] Assume that the pressure distribution function is P(x, y, t), where (x, y) represents the coordinates of any position on the contact surface and t represents time. A two-dimensional interpolation algorithm, such as bilinear interpolation or adaptive spline interpolation, can be used to calculate the pressure value P of each sensor in the pressure signal set P. i (t) and its position coordinates (x i ,y i ), and fit the pressure distribution function P(x, y, t), that is:
[0137] P(x,y,t)=f(P1(t),P2(t),...,P n (t);x1,y1,x2,y2,...,x n ,y n )
[0138] Where f represents the interpolation algorithm. This pressure distribution model can be used to analyze the pressure characteristics of the orthosis on the patient's skin in different postures, providing a basis for the subsequent optimization of the airbag pressure configuration. For example, local high-pressure areas can be identified based on the pressure distribution characteristics to adjust the pressure of the corresponding airbag in a targeted manner.
[0139] Step S30: establishing a vibration model of the orthosis according to the vibration signal set and the position of each vibration sensor, so as to evaluate the stability and comfort of the orthosis.
[0140] The specific implementation is as follows:
[0141] Vibration analysis techniques such as modal analysis can be used to analyze the vibration values V of each sensor in the vibration signal set V. j (t) and its position coordinates (x j ,y j ), establish the vibration model of the orthosis. The vibration model can be expressed as:
[0142]
[0143] Among them, M is the mass matrix, C is the damping matrix, K is the stiffness matrix, x(t) is the displacement vector, is the velocity vector, is the acceleration vector, and F(t) is the external load vector. By solving the vibration equation, the vibration characteristics of the orthosis at each vibration sensor position, such as amplitude, frequency, etc., can be obtained.
[0144] This vibration model can be used to evaluate the stability and wearing comfort of the orthosis. For example, the high amplitude area near the resonant frequency can be identified based on the vibration characteristics, and the airbag pressure configuration can be adjusted to suppress these vibrations, thereby improving the stability of the orthosis and the comfort of the user.
[0145] Step S40: establishing an airbag pressure distribution model according to the air pressure signal set and the position of each airbag, so as to adjust the pressure in the airbag.
[0146] The specific implementation is as follows:
[0147] Assume that the pressure distribution function is G(x, y, t), where (x, y) represents the airbag position coordinates and t represents time. The two-dimensional interpolation algorithm can also be used to calculate the pressure value G of each sensor in the pressure signal set G. l (t) and its position coordinates (x l ,y l ), and fit the pressure distribution function G(x, y, t), that is:
[0148] G(x,y,t)=g(G1(t),G2(t),…,G k (t);x1,y1,x2,y2,…,x k ,y k )
[0149] Here, g represents the interpolation algorithm. The air pressure distribution model provides a basis for adjusting the pressure in each airbag. For example, the area with low pressure can be identified based on the pressure distribution characteristics, and the pressure of the corresponding airbag can be appropriately increased, or the area with too high pressure can be identified and the pressure of the corresponding airbag can be appropriately reduced to achieve the goal of overall pressure optimization.
[0150] Step S50: Obtain the pressure distribution characteristics, vibration characteristics and air pressure distribution characteristics of the pressure distribution model, vibration model and airbag pressure distribution model under different postures of the user, and establish a posture-feature mapping relationship model to predict the orthopedic effect under different postures.
[0151] The specific implementation is as follows:
[0152] 1. The accelerometer or gyroscope installed on the outer surface of the orthosis can collect the motion signals of the user in different postures, and use pattern recognition algorithms, such as principal component analysis or Gaussian mixture model, to identify different posture features. Suppose the identified posture feature vector is a(t) = [a1(t), a2(t), ..., a p (t)] T , where p is the dimension of the posture feature.
[0153] 2. Combine the pressure distribution model P(x, y, t) and vibration model established in steps S20-S40 And the air pressure distribution model G(x, y, t), the characteristic indicators of each model in different postures can be extracted, such as:
[0154] Pressure distribution characteristic vector p(t) = [p1(t), p2(t), ..., p q (t)] T , where q is the dimension of the pressure characteristic
[0155] Vibration eigenvector v(t) = [v1(t), v2(t), ..., v r (t)] T , where r is the dimension of the vibrational feature
[0156] Air pressure distribution characteristic vector g(t) = [g1(t), g2(t), ..., g s (t)] T , where s is the dimension of the pressure feature
[0157] 3. Next, supervised learning algorithms, such as multivariate linear regression or neural networks, can be used to build a posture-feature mapping relationship model:
[0158] p(t)=f p (a(t))
[0159] v(t)=f v (a(t))
[0160] g(t)=f g (a(t))
[0161] Among them, f p , f v , f g is the mapping function. The model can predict the corresponding pressure distribution characteristics p(t), vibration characteristics v(t) and air pressure distribution characteristics g(t) according to the user's posture a(t) detected in real time, thereby predicting the correction effect under different postures.
[0162] Step S60: Establish a multi-objective optimization model with the goals of maximizing the correction effect, optimizing the wearing comfort and minimizing the power consumption, using the airbag pressure as the decision variable, and using the pressure distribution, vibration and air pressure safety thresholds as constraints.
[0163] The specific implementation is as follows:
[0164] 1. Take the correction effect F1, wearing comfort F2 and power consumption F3 as the optimization objective function. Among them:
[0165] The correction effect F1 can be characterized by the weighted average of the pressure distribution, that is, F1 = ∫∫w(x, y)P(x, y, t)dxdy, where w(x, y) is the weight function
[0166] The wearing comfort F2 can be represented by the weighted average value of vibration acceleration, that is, Where u(x, y) is the weight function
[0167] The power consumption F3 can be represented by the weighted average value of the airbag pressure, that is, where v l is the weight coefficient
[0168] 2. Set the pressure value of each airbag G l (t), l = 1, 2, ..., k as decision variables. The pressure distribution, vibration characteristics and air pressure value must meet certain safety threshold constraints, for example:
[0169] The maximum value of the pressure distribution does not exceed P max =40kPa
[0170] The maximum value of vibration acceleration does not exceed
[0171] The maximum airbag pressure does not exceed G max =80kPa
[0172] Based on the above, the following multi-objective optimization model can be established:
[0173] min{-F1, -F2, F3}
[0174] stP(x,y,t)≤P max
[0175]
[0176] G l (t)≤G max , l = 1, 2, ..., k
[0177] The decision variable is the pressure value G of each airbag. l (t). A multi-objective optimization algorithm, such as a genetic algorithm or a particle swarm optimization algorithm, can be used to solve the optimization model and obtain the optimal airbag pressure configuration scheme under different postures. This scheme can take into account the optimization of correction effect, wearing comfort and power consumption.
[0178] Step S70: Solve the multi-objective optimization model to obtain the optimal airbag pressure configuration scheme under different postures.
[0179] The specific implementation is as follows:
[0180] 1. Using a multi-objective optimization algorithm, such as a genetic algorithm or a particle swarm optimization algorithm, to solve the multi-objective optimization model established in step S60.
[0181] 2. The goal of the algorithm is to maximize the correction effect F1, optimize the wearing comfort F2, and minimize the power consumption F3 while satisfying the pressure distribution, vibration characteristics and air pressure safety constraints.
[0182] 3. Through iterative optimization, the optimal airbag pressure configuration scheme under different postures can be obtained This solution provides the best reference for the subsequent dynamic adjustment of the airbag pressure.
[0183] Step S80: According to the optimal airbag pressure configuration scheme, combined with the user's posture detected in real time, the pressure of each airbag is dynamically adjusted to achieve adaptive control of the orthotic force of the orthosis.
[0184] The specific implementation is as follows:
[0185] 1. Monitor the user's posture changes in real time, and predict the corresponding pressure distribution feature vector p(t), vibration feature vector v(t) and air pressure distribution feature vector g(t) through the posture-feature mapping relationship model established in step S50.
[0186] 2. Compare the prediction result with the optimal airbag pressure configuration solution obtained in step S70 Compare and determine the airbag and its pressure value that need to be adjusted. The specific adjustment strategy is as follows:
[0187]
[0188] Among them, K p , K d is the proportional and differential adjustment coefficient, and Δt is the adjustment period. This formula adopts the proportional-differential control strategy, which can realize the gradual adjustment of the airbag pressure and avoid the discomfort caused by sudden pressure changes.
[0189] 3. At the same time, a safety monitoring mechanism is set up. When the maximum value of the pressure distribution max{P(x, y, t)} exceeds P max =40kPa, or the maximum value of vibration acceleration Exceed When the device is in use, immediately perform airbag decompression to ensure safe use.
[0190] 4. Adaptive control of the orthotic force of the orthosis is achieved by dynamically adjusting the airbag pressure to improve the correction effect F1, enhance wearing comfort F2, and reduce power consumption F3.
[0191] In summary, the patent claims describe an adaptive control method for scoliosis braces based on airbag pressure. A multi-objective optimization model was established by modeling and analyzing pressure, vibration, and air pressure signals, and dynamic airbag pressure regulation was achieved in combination with real-time posture detection, thereby improving the correction effect, wearing comfort, and power efficiency. This method integrates multidisciplinary knowledge, including sensing technology, signal processing, mechanical vibration, optimization algorithms, etc., and provides an innovative solution for the intelligent control of scoliosis braces.
[0192] The entire specific implementation method involves a large number of mathematical formulas and algorithm principles, including:
[0193] 1. Use the two-dimensional interpolation algorithm to construct the pressure distribution model P(x, y, t) and the air pressure distribution model G(x, y, t).
[0194] 2. Use vibration analysis technology to establish the vibration model of the orthosis
[0195] 3. Use supervised learning algorithm to establish posture-feature mapping relationship model p(t)=f p (a(t)), v(t) = f v (a(t)), g(t) = f g (a(t)).
[0196] 4. Use multi-objective optimization algorithm to solve the multi-objective optimization model of pressure distribution, vibration characteristics and air pressure.
[0197] 5. Use proportional-differential control strategy to achieve dynamic regulation of airbag pressure.
[0198] Specifically, the principle of the present invention is to establish a pressure distribution model, a vibration model and an air pressure distribution model, combine the user's real-time posture information, and adopt a multi-objective optimization method to dynamically adjust the pressure of each airbag, thereby achieving coordinated optimization of correction effect, wearing comfort and power consumption. This idea conforms to the following principles and logic:
[0199] 1. The establishment of the pressure distribution model provides a basis for accurate control of the correction force. The scoliosis orthosis corrects spinal deformity by applying a certain correction force. The magnitude and distribution of the correction force directly determine the correction effect. Therefore, accurately describing the pressure distribution characteristics on the contact surface between the orthosis and the skin is the basis for achieving accurate correction control. The present invention adopts an interpolation algorithm to construct a pressure distribution model based on the signals of multiple pressure sensors on the contact surface, which can predict the pressure value at any position and provide a basis for airbag pressure regulation.
[0200] 2. The establishment of a vibration model helps to improve the comfort of use. In addition to the corrective force, the orthosis will also generate a certain amount of vibration during use. Excessive vibration will cause discomfort to the user and reduce the wearing comfort. The present invention establishes a vibration model of the orthosis through vibration analysis technology, which can predict the vibration characteristics of each area, and adjust the airbag pressure in a targeted manner to suppress local high vibration, thereby improving the overall comfort of use.
[0201] 3. The establishment of the air pressure distribution model lays the foundation for power optimization. The airbag pressure adjustment of the orthosis directly affects the power consumption. The present invention also establishes an air pressure distribution model, which can accurately predict the pressure state of each airbag, provide a basis for multi-objective optimization, and minimize the overall power consumption while meeting the correction effect and comfort requirements.
[0202] 4. Establishment of posture-feature mapping model to achieve adaptive control. When the human body is in different postures, the shape and force conditions of the spine will change, which will in turn affect the pressure distribution, vibration characteristics and air pressure distribution of the orthosis. The present invention establishes a posture-feature mapping model by real-time monitoring of the user's posture changes, which can predict the pressure, vibration and air pressure characteristics under different postures, provide a basis for dynamically adjusting the airbag pressure, and achieve adaptive control.
[0203] 5. Solving multi-objective optimization, taking into account correction effect, comfort and power consumption. The present invention sets the maximization of correction effect, optimization of wearing comfort and minimization of power consumption as a multi-objective optimization problem. By solving the multi-objective optimization algorithm, the optimal airbag pressure configuration scheme under different postures is obtained, which can not only meet the correction requirements, but also improve comfort and reduce power consumption, reflecting the systematic optimization design idea.
[0204] In summary, the technical solution of the adaptive control method of the present invention closely revolves around the key performance indicators of the scoliosis orthosis, fully utilizes advanced sensing technology, signal processing and optimization algorithms, establishes multiple key models, and integrates them into the dynamic airbag pressure regulation, thereby achieving coordinated optimization of correction effect, wearing comfort and power consumption, which meets the technical requirements for the intelligent development of scoliosis orthosis.
[0205] In order to better understand the present invention, an embodiment of a specific application scenario of the present invention is provided below:
[0206] The orthopedics department of a certain hospital diagnosed and treated a 15-year-old female patient named Xiaohong for scoliosis. After a preliminary examination, the doctor confirmed that Xiaohong suffered from mild scoliosis, with a vertebral rotation angle of about 20°. After communicating with Xiaohong and her family, it was decided to use a scoliosis orthosis for conservative treatment.
[0207] The doctor customized a Cheneau-style orthosis for Xiaohong based on her height, weight and other basic information. The orthosis is made of polypropylene material and has a total of 5 adjustable airbags distributed in the thoracic, lumbar and pelvic regions of the orthosis. In order to achieve intelligent control of the orthosis, the doctor decided to adopt the adaptive control method based on airbag pressure proposed in the present invention.
[0208] 1. Collection of pressure, vibration and air pressure signals
[0209] First, during the orthosis production process, the doctor placed sensors in the following locations:
[0210] (1) A total of 12 pressure sensors are placed on the contact surface between the orthosis and the skin, distributed in the thoracic, lumbar and pelvic regions. The position coordinate information of each pressure sensor is recorded on the design drawing of the orthosis.
[0211] (2) A total of 8 vibration sensors are placed on the outer surface of the orthosis, distributed in key locations. The position coordinates of each vibration sensor are also recorded.
[0212] (3) A pressure sensor is installed at the vent of each of the five airbags to monitor the pressure changes inside the airbag in real time.
[0213] In addition, the doctor also installed an integrated inertial measurement unit (IMU) on the outer surface of the orthosis, which includes a three-axis accelerometer and a three-axis gyroscope to detect the patient's movement posture in real time.
[0214] While Xiaohong was wearing the orthosis, each sensor collected pressure signals, vibration signals, and air pressure signals in real time, and transmitted these raw data to a nearby control terminal device via Bluetooth for storage and analysis. At the same time, the IMU also monitored Xiaohong's posture changes in different movements in real time. After 10 minutes of data collection, the doctor obtained a preliminary set of pressure signals P, vibration signals V, and air pressure signals G, as well as the corresponding posture features a(t).
[0215] 2. Establishment of pressure distribution model
[0216] Based on the data of the 12 pressure sensors recorded in the pressure signal set P and their specific position coordinates on the orthosis, the doctor used a bilinear interpolation algorithm to construct a pressure distribution model P(x, y, t) of the contact surface between the orthosis and the skin. This model can predict the pressure value at any position (x, y) on the contact surface at time t.
[0217] The doctor conducted a preliminary analysis of the pressure distribution model and found that when Xiaohong stood upright, there were two local high-pressure areas in the thoracic and lumbar regions of the orthosis, with pressure values reaching about 35kPa, while the pressure in the pelvic region was relatively low, about 20kPa. This shows that the pressure distribution of the orthosis needs to be further optimized and adjusted to achieve a more uniform state.
[0218] 3. Establishment of vibration model
[0219] Based on the 8 vibration sensor data recorded in the vibration signal set V and their position coordinates on the orthosis, the doctor used the modal analysis method to establish a vibration model of the orthosis. This model describes the vibration characteristics of the orthosis at different positions, including amplitude and frequency.
[0220] The analysis results show that when Xiaohong walks slowly, the local vibration of the orthosis is quite intense, especially in the thoracic spine area, where the vibration acceleration reaches up to 4.5m / s^2. These high-vibration areas may have an adverse effect on the user's wearing comfort.
[0221] 4. Establishment of air pressure distribution model
[0222] The doctor further analyzed the data of the five air pressure sensors recorded in the air pressure signal set G, and combined with their position information on the orthosis, constructed the airbag pressure distribution model G(x, y, t) using a bilinear interpolation algorithm. This model can predict the pressure value of any airbag position (x, y) at time t.
[0223] Preliminary results show that when Xiaohong stands upright, the pressure distribution of the five airbags is relatively uniform, about 60kPa. But when Xiaohong walks slowly, the pressure of some airbags fluctuates significantly, reaching as high as 75kPa, while the pressure of other airbags is low, only about 50kPa. This uneven pressure distribution may reduce the stability and wearing comfort of the orthosis.
[0224] 5. Establishment of posture-feature mapping relationship model
[0225] The doctor further analyzed the posture data a(t) collected by the IMU and found that the pressure distribution model P(x, y, t) and the vibration model P(x, y, t) were different in different postures, such as standing upright, walking slowly, and walking uphill. And the air pressure distribution model G(x, y, t) have changed significantly.
[0226] In order to achieve adaptive control, the doctor used a multivariate linear regression algorithm to establish a posture-feature mapping relationship model:
[0227] p(t)=f p (a(t))
[0228] v(t)=f v (a(t))
[0229] g(t)=f g (a(t))
[0230] Through this model, doctors can predict the corresponding pressure distribution characteristics p(t), vibration characteristics v(t) and air pressure distribution characteristics g(t) based on the user's posture a(t) detected in real time, providing a basis for subsequent airbag pressure optimization.
[0231] 6. Establishment and solution of multi-objective optimization model
[0232] With the above models as a basis, the doctor then established a multi-objective optimization model, taking maximizing the correction effect, optimizing wearing comfort and minimizing power consumption as the three objective functions:
[0233] min{-F1, -F2, F3}
[0234] stP(x,y,t)≤P max =40kPa
[0235]
[0236] G l (t)≤G max =80kPa, l=1, 2,...,5
[0237] Among them, the correction effect F1 is represented by the weighted average of pressure distribution, the wearing comfort F2 is represented by the weighted average of vibration acceleration, and the power consumption F3 is represented by the weighted average of airbag pressure. The pressure value G of the 5 airbags l (t) is used as a decision variable and is constrained by the pressure distribution, vibration characteristics, and the safety threshold of air pressure.
[0238] To solve this multi-objective optimization problem, the doctor selected a genetic algorithm. Through iterative optimization, the optimal airbag pressure configuration scheme under different postures was finally obtained.
[0239] 7. Adaptive adjustment of dynamic airbag pressure
[0240] With the optimal airbag pressure configuration scheme as a reference, the doctor then realized the dynamic regulation and control of the orthosis airbag pressure. The specific process is as follows:
[0241] (1) Monitor Xiaohong's movement posture changes a(t) in real time, and predict the corresponding pressure distribution characteristics p(t), vibration characteristics v(t), and air pressure distribution characteristics g(t) through the posture-feature mapping relationship model.
[0242] (2) Compare the prediction results with the optimal airbag pressure configuration plan Compare and determine the airbag and its pressure value that need to be adjusted. Use the following proportional-differential control strategy to adjust the pressure:
[0243]
[0244] Among them, Δt is the adjustment period, K p , K d is the proportional and differential adjustment coefficient. This strategy can achieve gradual adjustment of the airbag pressure and avoid discomfort caused by sudden changes.
[0245] (3) At the same time, a safety monitoring mechanism is set up. When the maximum value of the pressure distribution max{P(x, y, t)} exceeds P max =40kPa, or the maximum value of vibration acceleration Exceed When the device is in use, immediately perform airbag decompression to ensure safe use.
[0246] Through the above-mentioned dynamic adjustment strategy, the orthosis can adaptively adjust the pressure configuration of each airbag according to Xiaohong's real-time movement posture, which not only meets the needs of correction effect, but also improves wearing comfort and reduces power consumption.
[0247] 8. Analysis of clinical application effects
[0248] During the three-month treatment, the doctor closely monitored Xiaohong's spinal deformity correction. Through regular follow-up examinations with X-rays, it was found that Xiaohong's vertebral rotation angle was gradually corrected from the initial 20° to 12°, close to the normal level. At the same time, Xiaohong said that she felt more comfortable when using the orthosis and could wear it for a long time without feeling obvious fatigue. The doctor measured the energy consumption of the orthosis system and found that compared with the traditional manual adjustment method, the power consumption was reduced by about 30%.
[0249] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.
Claims
1. A method for controlling the corrective force of a scoliosis orthosis with an airbag, characterized in that: The following steps are involved: S10, acquiring pressure signals, vibration signals, and air pressure signals collected by multiple pressure sensors, multiple vibration sensors, and multiple air pressure sensors to form a pressure signal set, a vibration signal set, and an air pressure signal set; S20, establishing a pressure distribution model of the contact surface between the orthosis and the skin according to the pressure signal set and the position of each pressure sensor; S30, establishing a vibration model of the orthosis according to the vibration signal set and the position of each vibration sensor, so as to evaluate the stability and comfort of the orthosis; S40, establishing an airbag pressure distribution model according to the air pressure signal set and the position of each airbag, so as to adjust the pressure in the airbag; S50, obtaining pressure distribution characteristics, vibration characteristics and air pressure distribution characteristics of the pressure distribution model, vibration model and airbag pressure distribution model under different postures of the user, and establishing a posture-feature mapping relationship model for predicting the orthopedic effect under different postures; S60, establishing a multi-objective optimization model, with the objectives of maximizing the correction effect, optimizing the wearing comfort and minimizing the power consumption, with the airbag pressure as the decision variable, and with the pressure distribution, vibration and air pressure safety threshold as the constraint conditions; S70, solving the multi-objective optimization model to obtain the optimal airbag pressure configuration scheme under different postures; S80, dynamically adjusting the pressure of each airbag according to the optimal airbag pressure configuration scheme and in combination with the user's posture detected in real time, so as to achieve adaptive control of the orthotic force of the orthosis.
2. The airbag control method of the scoliosis orthosis force according to claim 1, characterized in that: The multiple pressure sensors are multiple pressure sensors arranged on the contact surface between the orthosis and the skin.
3. The airbag control method of the scoliosis orthosis force according to claim 1, characterized in that: The plurality of vibration sensors are a plurality of vibration sensors arranged on the outer surface of the orthosis.
4. The airbag control method of the scoliosis orthosis force according to claim 1, characterized in that: The air pressure sensor is arranged at the vent of the airbag and is used to collect the air pressure in the airbag.
5. The airbag control method of the scoliosis orthosis force according to claim 1, characterized in that: The user's posture is detected and acquired by an acceleration sensor or a gyroscope arranged on the outer surface of the orthosis.
6. The airbag control method of the scoliosis orthosis force according to claim 1, characterized in that: The posture-feature mapping relationship model adopts a multivariate linear regression model.
7. The airbag control method of the scoliosis orthosis force according to claim 1, characterized in that: The method used to solve the multi-objective optimization model is a genetic algorithm.
8. The airbag control method of the scoliosis orthosis force according to claim 1, characterized in that: When a genetic algorithm is used to solve the multi-objective optimization model, the fitness function is a similarity function of a vector composed of each objective.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program instructions, and when the program instructions are executed, they are used to execute the airbag control method for the corrective force of a scoliosis orthosis according to any one of claims 1 to 8.
10. An airbag control system for the corrective force of a scoliosis orthosis, characterized in that: A computer-readable storage medium comprising the computer-readable storage medium of claim 9.
Citation Information
Patent Citations
Correction method for intelligent correction brace
CN111096837A
Intelligent correction brace and correction method thereof
CN111096838A
Intelligent orthopedic device and processing method thereof
CN111110419A
Scoliosis orthosis based on pressure control and using method of scoliosis orthosis
CN111588528A
Scoliosis orthosis based on artificial intelligence and control method thereof
CN112168448A
