WO2014194609A1 - 一种基于肌电信号和传感器信号实现精细实时运动的控制方法 - Google Patents

一种基于肌电信号和传感器信号实现精细实时运动的控制方法 Download PDF

Info

Publication number
WO2014194609A1
WO2014194609A1 PCT/CN2013/087570 CN2013087570W WO2014194609A1 WO 2014194609 A1 WO2014194609 A1 WO 2014194609A1 CN 2013087570 W CN2013087570 W CN 2013087570W WO 2014194609 A1 WO2014194609 A1 WO 2014194609A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
sensor
stimulation pulse
stimulation
control method
Prior art date
Application number
PCT/CN2013/087570
Other languages
English (en)
French (fr)
Inventor
王志功
宗思浩
吕晓迎
徐建
王苏阳
Original Assignee
南京神桥医疗器械有限公司
东南大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 南京神桥医疗器械有限公司, 东南大学 filed Critical 南京神桥医疗器械有限公司
Publication of WO2014194609A1 publication Critical patent/WO2014194609A1/zh

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/1107—Measuring contraction of parts of the body, e.g. organ, muscle
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
    • A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
    • A61F2/50—Prostheses not implantable in the body
    • A61F2/68—Operating or control means
    • A61F2/70—Operating or control means electrical
    • A61F2/72—Bioelectric control, e.g. myoelectric
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
    • A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
    • A61F2/50—Prostheses not implantable in the body
    • A61F2/54—Artificial arms or hands or parts thereof
    • A61F2/58—Elbows; Wrists ; Other joints; Hands
    • A61F2/583—Hands; Wrist joints
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
    • A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
    • A61F2/50—Prostheses not implantable in the body
    • A61F2/76—Means for assembling, fitting or testing prostheses, e.g. for measuring or balancing, e.g. alignment means
    • A61F2002/7615—Measuring means

Definitions

  • the present invention is a method for realizing fine-to-finger real-time motion control based on a myoelectric signal and a curvature sensor, and belongs to the field of cross-technology of electronic science and clinical medicine.
  • the reconstruction device comprises a detection electrode array, a signal processing circuit, a communication channel, an excitation signal generation circuit and a stimulation electrode array.
  • the detecting electrode array adopts a single-ended form of a plurality of detecting electrodes and a common reference electrode, or a differential form of pairwise pairing.
  • This device can be used in any occasion where reconstruction of the motor function is required, including the reconstruction of the motor function of the patient with spinal cord injury, the reconstruction of the motor function of the hemiplegia caused by the stroke, and the training of the healthy person.
  • the above reconstruction device can transmit the multi-channel myoelectric signal generated when the collected healthy limb moves, through the signal a functional electrical excitation system that is transmitted to the limbs by wired or wireless communication, and generates a multi-channel stimulation electrical signal by using a stimulation waveform generation algorithm, applied to the stimulation electrode array worn on the limbs, and generates an electromyographic signal. Collecting similar harmonious movements of the limbs, achieving the purpose of reconstructing the limb's movement function.
  • the myoelectric signal is an analog signal that is easily disturbed, it is highly demanding in the environment. In a high-noise environment, other noises are easily mixed, making the motion recognition result inaccurate. Summary of the invention
  • the present invention is directed to the deficiencies of the above problems, and proposes a control method for realizing fine real-time motion based on the myoelectric signal and the sensor signal, and uses the curvature sensor to collect the curvature information of each limb movement of the main controller, and simultaneously detects through the body surface.
  • the electrode collects the corresponding electromyogram signal generated by the main controller at the moment, causing the corresponding electromyographic signal generated on the relevant muscle, and generates the stimulation pulse signal by using the stimulation pulse generation algorithm; and using the curvature information as a channel trigger signal of the stimulation pulse signal of the muscle;
  • the accused uses the myoelectric stimulation electrode to apply the obtained stimulation pulse signal to the corresponding muscles, so as to urge the controlled subject to move according to the curvature information of the main control member's limb, realizing fine-to-finger real-time motion control, and achieving accurate movement. Copy and rebuild.
  • a control method for realizing fine real-time motion based on a myoelectric signal and a sensor signal comprising the following steps: Step one, using a curvature sensor to collect each limb of the master controller The bending information of the action, and at the same time, the body surface detecting electrode collects the corresponding electromyogram signal generated by the main controller at the moment, and the corresponding electromyography signal generated by the relevant muscle; Step 2, through the functional electric excitation system, the functional electric excitation system integrates the stimulation pulse generating algorithm The stimulation pulse generation algorithm generates a stimulation pulse signal by using the myoelectric signal corresponding to the curvature information in step 1.
  • the functional electrical excitation system uses the curvature information described in step 1 as a channel for the stimulation pulse signal corresponding to the limb motion.
  • the stimulation pulse generation algorithm refers to an algorithm for generating a stimulation signal whose amplitude and frequency are both proportional to the collected electrical signal.
  • the myoelectric signal refers to an electromyogram signal processed by an amplification, filtering, and A/D conversion circuit
  • the curvature information refers to curvature information processed by the A/D conversion circuit
  • the method for determining the channel trigger signal is: first setting a threshold value, when the curve sensor collects a signal below a threshold value, it is considered that bending occurs, and then pulse stimulation is selected for the corresponding output channel. .
  • the curvature sensor when the action performed by the master is finger movement, the curvature sensor is disposed on the back of the five fingers; the hand index of the collection of the electromyogram signal is at least two, and the stimulation pulse signal can pass through the muscle Electrical signal fitting and then stimulation Pulse generation algorithm generation.
  • the stimulation pulse generation algorithm is integrated on the MSP430 series micro-processing single chip microcomputer.
  • a control method for realizing fine real-time motion based on the myoelectric signal and the sensor signal has the following beneficial effects:
  • the bending degree sensor is used to collect the curvature of each limb movement of the main controller.
  • Information at the same time, through the body surface detecting electrode to collect the corresponding electromyographic signal generated by the main controller at the moment, causing the corresponding electromyographic signal generated on the relevant muscle, and generating the stimulation pulse signal through the stimulation pulse generation algorithm; using the curvature information as the stimulation pulse of the muscle The channel trigger signal of the signal; then the controlled person applies the obtained stimulation pulse signal to the corresponding muscle through the myoelectric stimulation electrode, so that the controlled subject moves according to the curvature information of the main control member's limb, realizing the fine to finger real time.
  • the motion control achieves accurate copying and reconstruction of the motion, so the present invention adds sensor signals as an aid to improve the accuracy of identifying fine motions, especially the fine movements of the hand.
  • the EMG signal is mainly used, and the sensor signal is supplemented.
  • the acquired EMG signal is used to generate the relevant stimulation pulse sequence through the stimulation pulse generation algorithm, so that the output stimulation pulse is related to the original EMG signal;
  • the sensor signal is used as a channel trigger signal, and the limb motion curvature information is collected to determine which limbs are moving, to select a corresponding output stimulation limb, and the stimulation pulse is output to the corresponding muscle site of the controlled person to perform different muscle nerves.
  • the stimulus corresponding to the intensity of the master's action achieves fine real-time control.
  • the advantage of this is that it solves the shortcomings that the EMG signal can not recognize the fine action.
  • the EMG signal in the system compensates for the problem that the sensor signal cannot collect the exercise intensity.
  • the information fusion can successfully complete the action and Intensity recording, restores a complete action feature record, and can copy it to achieve fine, real-time motion control.
  • FIG. 1 is a schematic diagram of a system for implementing a fine-to-finger real-time motion control method based on a myoelectric signal and a sensor signal.
  • FIG. 2 is a schematic view showing the distribution of a curvature sensor on a hand.
  • FIG. 3 is a circuit diagram of a signal acquired by a curvature sensor.
  • a control method for realizing fine real-time motion based on the myoelectric signal and the sensor signal in the embodiment, as shown in FIG. 1-3, includes the following steps: In the first step, the curvature sensor is used to collect the curvature information of each limb's limb movements, and at the same time, the body surface detection electrode is used to collect the corresponding electromyography signal generated by the main muscle of the main controller at the moment;
  • the functional electro-excitation system is integrated with a stimulation pulse generation algorithm, and the stimulation pulse generation algorithm generates a stimulation pulse signal according to the myoelectric signal corresponding to the curvature information in step 1.
  • the excitation system uses the curvature information described in step 1 as a channel trigger signal corresponding to the stimulation pulse signal of the limb motion; the electromyography signal is processed by an amplification, filtering, A/D conversion circuit; the stimulation pulse generation algorithm is Refers to the stimulation signal that produces the amplitude and frequency of the EMG signal in proportion to the acquired EMG signal.
  • the curvature information is used as a channel trigger signal of the stimulation signal of the muscle, and the curvature information refers to the curvature information processed by the A/D conversion circuit.
  • the channel trigger signal is used to select a corresponding output stimulation channel of the stimulation signal; the channel trigger signal is selected by: first setting a threshold, and when the curvature sensor collects the signal below the threshold, the bending is considered to occur, and then Pulse stimulation is applied to the corresponding output channel selection.
  • the accused applies the stimulation pulse signal obtained in step 2 to the corresponding muscle through the myoelectric stimulation electrode, so as to prompt the controlled subject to move according to the curvature information of the main controller of the first step.
  • the bending sensor is disposed on the back of the five fingers; and the body detecting electrodes are respectively placed on the respective muscle detecting points of the finger.
  • the myoelectric stimulation electrode is placed on each muscle stimulation point of the finger of the controlled person; since there are five fingers, the myoelectric signal can be divided into five channels, which correspond to five fingers (thumb, index finger, Middle finger, ring finger and little finger), when the master's finger moves, the first step, that is, the bending sensor, collects the bending information of the five finger movements of the master, and simultaneously collects five masters at the moment through the surface detecting electrode.
  • the finger action causes a corresponding electromyogram signal generated on each finger muscle; and then, through the second step, generates a stimulation pulse signal by the electromyographic signal corresponding to the curvature information of each finger in step one through a functional electric excitation system, the function
  • the electric excitation system uses the curvature information of each finger as described in step 1 as a channel trigger signal corresponding to the stimulation pulse signal of the finger motion.
  • the selection method of the channel trigger signal is as follows: First, a threshold is set, and when the bending sensor on the finger collects the signal below the threshold, it is considered that the bending occurs, and then the pulse stimulation is selected on the corresponding finger output channel; After the third step, the controlled person applies the stimulation pulse signal obtained in step 2 to the corresponding finger muscle through the myoelectric stimulation electrode, so as to urge the controlled finger to act according to the curvature information of the master's finger in step 1.
  • the algorithm of the functional electrical excitation system monitors the corresponding When the sensor signal of the index finger changes, and the other sensor signals do not change, it can be judged that the action of the hand of the master is the bending of the index finger, and the corresponding stimulation pulse for bending the index finger of the controlled person can be generated.
  • the body surface detecting electrode can measure only the myoelectric signals generated by the two finger movements, that is, only two channels are measured.
  • the myoelectric signal, and the stimulation pulse signal can be generated by fitting the myoelectric signal and then by the stimulation pulse generation algorithm, and the action of the finger can be determined by the curvature sensor.
  • the principle of the invention A real-time motion control method based on the myoelectric signal and the sensor signal to achieve fine-to-finger, the core idea is: collecting the electromyogram signal on the muscle of the master's action-related muscle, through the stimulation pulse generation algorithm Generate relevant stimulation pulse sequences, and then use each finger curvature information collected by the curvature sensor as a channel trigger signal, select a corresponding output stimulation channel, and output the generated stimulation pulse sequence to the muscle stimulation site associated with the controlled person.
  • functional electrical stimulation related to the source nerve/electromyography signal for different target muscle nerves real-time control of fine action to fine fingers is realized.
  • the sensor signal is acquired by a curvature sensor.
  • the principle of the bending sensor is that the change in the degree of bending causes a change in the impedance of the self, thereby achieving the purpose of sensing the degree of bending.
  • the introduction of sensor signals is mainly to solve the problem of recognition of fine movements. It is possible to recognize some coarse movements such as grasping by relying solely on the myoelectric signal, but for the movement of the fingers, it is difficult to recognize the electromyographic signals alone, and the curvature sensor is introduced. It is possible to monitor the movement and bending of each finger to achieve the purpose of identifying fine movements.
  • the myoelectric signal is an electromyographic signal of a specific part of the human body collected by an amplification, filtering, A/D conversion circuit. After the acquisition, the signal processing circuit is sent. Generally, according to the complexity of the corresponding action, how many paths of the EMG signal are selected, and the two-way EMG signal is matched with the sensor signal to achieve a good effect.
  • the bending degree sensor signal is used as a switching signal, and the specific method is to set a threshold.
  • the bending sensor collects the signal below the threshold, the finger is considered to be bent, and then the corresponding output channel is pulse-stimulated to reach The purpose of the master controller's synchronization.
  • collecting the myoelectric signal on a specific muscle generates a related stimulation pulse sequence through a stimulation pulse generation algorithm, specifically by collecting an electromyogram signal on a specific muscle of the master, using an algorithm to generate amplitude and frequency and collecting the generated
  • the myoelectric signal is proportional to the stimulation pulse sequence, so that the output signal is directly proportional to the acquired host's myoelectric signal.
  • the master's motion amplitude is large, the generated myoelectric signal is large, and the output stimulation pulse is also Correspondingly, there is a good feedback correlation between the master and the controller.
  • Fine motion control accurate to the movement of the finger means that the multi-channel stimulation pulse generated by the sensor signal and the electromyogram signal is used to stimulate the specific muscle site of the controlled person, so that the corresponding object can be accurately moved.
  • the level of sophistication of the movement can reach the level of a single finger movement.
  • Real-time means that after the collector makes an action, the stimulated object within l-2s can make a corresponding fine action response.
  • Signal acquisition processing and algorithms are a very fast process, which takes time in milliseconds, and the main delay is generated by The response time of the muscle to the stimulation pulse, because the signal takes a certain amount of time to produce the energy that allows the muscle to move, especially in the case of relatively weak movements.
  • a larger amplitude action such as gripping has a shorter delay, and the motion of the stimulating object is almost synchronous with the motion of the collected object.
  • the stimulation pulse generation algorithm is implemented by a single-chip microcomputer. In the experiment, TI's MSP430 series microprocessor is used, and the same algorithm can be implemented by other series of microprocessors. After the algorithm processes, the generated stimulation pulse signal is input into the stimulation circuit, and the final stimulation pulse waveform is generated by the stimulation circuit, and the stimulation pulse is output to the sensitive control site of the corresponding moving muscle of the human body through the body surface electrode.
  • the method of implementing fine-to-finger motion control based on the myoelectric signal and the sensor can realize that the healthy person can drive the deaf patient to perform fine movements in real time, or the patient's healthy limbs can drive the limbs to perform the same fine action in real time, or the healthy person drives One or more healthy people perform real-time fine movements to achieve accurate reproduction and reconstruction of the movement, whether for the rehabilitation of the paralyzed patient or the operation training of the healthy person, these functions can produce positive help.
  • the invention is characterized by: introduction of muscle The idea that the electrical signal and the sensor signal are fused together, the electromyogram signal is generated to generate a corresponding stimulation pulse sequence, and the sensor signal is added as a switching signal to select an output channel.
  • the occurrence and intensity of the action can be recognized, and the result is output.
  • the EMG signal can not recognize the fine action.
  • the EMG signal in the system compensates for the problem that the sensor signal cannot collect the exercise intensity.
  • the information fusion can successfully complete the action occurrence and intensity. Record, restore a complete action feature record, and copy it out to achieve fine-to-finger real-time motion control.
  • the healthy person can drive the paralyzed patient to perform real-time fine movements, or the patient's healthy limbs can drive the limbs to perform the same fine movements in real time, or the healthy person can drive one or more healthy people to perform fine movements to achieve accurate movements.
  • Reproduction and reconstruction, whether for the rehabilitation of deaf patients or the training of healthy people, these functions can be positively helpful.
  • a specific embodiment of the present invention uses this method to implement an example in which a healthy person drives a deaf patient to perform real-time fine motion.
  • the electromyographic signal of the specific muscle of the master is collected by the body surface electrode, and converted into a digital signal by AD transformation.
  • the bending sensor signal is used to collect the motion information of the master's finger.
  • the distribution of the curvature sensor on the hand is shown in Figure 2.
  • the circuit diagram of the acquired signal is shown in Figure 3. Through this distribution, the curvature information of each finger can be monitored, thereby restoring the precise movement of the hand.
  • the circuit in Figure 4 is used to convert the analog signal collected by the sensor into a digital signal, where Rf represents the curvature sensor, and the sensor changes its impedance as the curvature changes, thereby changing the output voltage. After the conversion, the information of the curvature can be expressed by voltage.
  • AD After AD, it is sent to the MCU, in the MCU.
  • the algorithm will detect that the sensor signal corresponding to the index finger has changed, and the other sensor signals have not changed, it can be judged that the action of the master's hand is the bending of the index finger, and the corresponding index finger can be bent.
  • Stimulation pulse At the same time, the pulse stimulation generation algorithm in the MCU generates a stimulation pulse sequence of corresponding intensity according to the magnitude of the acquired myoelectric signal: when the myoelectric signal is strong, a strong stimulation pulse is generated, and when the myoelectric signal is weak, it is weak. Stimulus pulse.
  • the synchronization of the motion of the master and the controlled person can be well realized, so that the system can judge the type of the stimulus, and can judge the intensity of the stimulus, and achieve the purpose of real-time fine motion control.
  • the stimulation circuit mainly determines the magnitude of the stimulation pulse and the length of the pulse according to the key indicators in the previous algorithm, and can achieve the purpose of making a specific action by stimulating the corresponding position of the human body.
  • the master performs an action, and in a short period of time, the controlled person performs the same action, which may be performed for both the recovery of the paralyzed patient and the operation training of the healthy person. Positive help.
  • sensor signals are added here as an aid to improve the accuracy of identifying fine motions, especially the fine movements of the hand.
  • the EMG signal is the main, the sensor signal is supplemented, and an "AND" algorithm is operated on the two, and then the algorithm is used to stimulate the different muscle nerves corresponding to the control intensity. Real-time control of finger movements.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Veterinary Medicine (AREA)
  • Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Cardiology (AREA)
  • Vascular Medicine (AREA)
  • Transplantation (AREA)
  • Physiology (AREA)
  • Dentistry (AREA)
  • Physics & Mathematics (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Electrotherapy Devices (AREA)
  • Prostheses (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

本发明涉及一种基于肌电信号和传感器信号实现精细实时运动的控制方法，属于电子科学与临床医学的交叉技术领域。其核心思想在于：利用体表电极采集到的健康肢体动作时产生的肌电信号和弯曲度传感器采集到的各个手指弯曲程度信号，将肌电信号通过刺激脉冲生成算法生成刺激信号；将弯曲度信息作为该肌肉的刺激信号的通道触发信号；然后将生成刺激信号输出至被控者相关的肌肉刺激位点上，对相应肢体的特定部位进行刺激，实现精细、实时动作控制，达到动作的准确复制和重建。

Description

说明书
一种基于肌电信号和传感器信号实现精细实时运动的控制方法 技术领域
[0001] 本发明是一种基于肌电信号和弯曲度传感器实现精细到手指的实时运动控制方法， 属于电子科学与临床医学的交叉技术领域。
背景技术
[0002] 偏瘫病人肢体动作功能的丧失给他们自身及家庭乃至社会带来巨大的痛苦和负担。 瘫痪肢体动作功能的重建， 一直是康复医学重点研究的目标。 多年以来， 国内外的科学家 们为瘫痪病人设计了众多控制肢体动作的功能电刺激装置， 主要分为植入式的和非植入式 两大类。 植入式无可避免地为患者带来手术的困扰， 而非植入式的功能电刺激由于通道数 目少， 无法达到精确的控制。 重要的是， 无论是植入式的还是非植入式的刺激系统， 由于 对各种动作的神经及肌电编码的认知缺失， 只能通过人工构想的脉冲编码作为控制信号产 生简单重复的肢体动作， 这些动作和正常人肢体动作的自由度和和谐度相去甚远。 由于健 康肢体肌电信号真实反映着实时的肌肉活动方式， 利用健康肢体动作时产生的肌电信号作 为再生的刺激信号， 就弥补了现有功能电刺激系统的缺陷。
[0003] 在 ZL 200510135541.6的发明专利中， 提出了 "微电子系统辅助神经信道功能恢复 方法及其装置"， 用于受损脊髓神经的信道桥接、 信号再生和功能重建。 这一专利要处理 的是动作和感觉的神经电位脉冲序列 （编码）。 该发明专利的特征是： 1 ) 应用目标是同体 受损的脊髓神经， 2) 采用近距离有线的生物神经 -电子接口， 3) 装置需要手术植入 在专利号为 201210342507.6，公开了一种基于肌电信号通信机理的瘫痪肢体功能重建方法 及其装置的发明专利中， 该重建装置包括探测电极阵列、 信号处理电路、 通信信道、 激励 信号生成电路和刺激电极阵列。 所述的探测电极阵列采用多个探测电极和一个共用参考电 极的单端形式，或采用两两配对的差动形式。此装置可用于任何需要动作功能重建的场合， 包括脊髓损伤的瘫痪病人动作功能重建， 中风导致的偏瘫病人动作功能重建， 健康人的动 作训练等等。
[0004] 上述重建装置可以将采集到的健康肢体动作时产生的多通道肌电信号， 经过信号处 理， 以有线或无线的通信方式传送至瘫痪肢体上的功能电激励系统， 利用刺激波形生成算 法生成多通道的刺激电信号， 施加于穿戴在瘫痪肢体上的刺激电极阵列， 产生与肌电信号 采集肢体相似的和谐动作， 达到瘫痪肢体动作功能重建的目的。 但是由于肌电信号采集和 处理的复杂性， 对于精细动作， 尤其是手部的精细动作， 很难将相应的肌电信号进行多路 分离， 影响了判断动作的准确性。 而且由于肌电信号是一个很容易受干扰的模拟信号， 对 环境要求很高， 在高噪声的环境中， 容易混入其他噪声， 使得动作识别结果不准确。 发明内容
[0005] 本发明针对上述问题的不足，提出一种基于肌电信号和传感器信号实现精细实时运 动的控制方法， 采用弯曲度传感器采集主控者各肢体动作的弯曲度信息， 同时通过体表探 测电极采集此刻主控者肢体动作致使相关肌肉上产生的对应肌电信号， 将肌电信号通过刺 激脉冲生成算法生成刺激脉冲信号； 将弯曲度信息作为该肌肉的刺激脉冲信号的通道触发 信号； 然后被控者通过肌电刺激电极， 对相应的肌肉施加获得的刺激脉冲信号， 以促使被 控者肢体按照主控者肢体的弯曲度信息动作， 实现精细到手指的实时动作控制， 达到动作 的准确复制和重建。
[0006] 本发明为解决上述技术问题提出的技术方案是： 一种基于肌电信号和传感器信号实 现精细实时运动的控制方法， 包括以下步骤： 步骤一， 采用弯曲度传感器采集主控者各肢 体动作的弯曲度信息， 同时通过体表探测电极采集此刻主控者肢体动作致使相关肌肉上产 生的对应肌电信号； 步骤二， 通过功能电激励系统， 该功能电激励系统集成有刺激脉冲生 成算法， 所述刺激脉冲生成算法将步骤一所述弯曲度信息对应的肌电信号生成刺激脉冲信 号， 所述功能电激励系统通过步骤一所述的弯曲度信息作为对应肢体动作的刺激脉冲信号 的通道触发信号； 步骤三、 被控者通过肌电刺激电极， 对相应的肌肉施加步骤二获得的刺 激脉冲信号， 以促使被控者肢体按照步骤一所述主控者肢体的弯曲度信息动作。
[0007] 优选的： 所述刺激脉冲生成算法是指将肌电信号生成幅度、 频率都与采集的电信号 成正比的刺激信号的算法。
[0008] 优选的： 所述肌电信号是指通过放大、 滤波、 A/D转换电路处理过的肌电信号； 所 述弯曲度信息是指通过 A/D转换电路处理过的弯曲度信息。
[0009] 优选的： 所述通道触发信号的判断方法为： 首先设定一个阈值， 当弯曲度传感器收 集信号低于阈值时， 就认为发生了弯曲， 进而在对相应的输出通道选择进行脉冲刺激。
[0010] 优选的： 当主控者所做的动作为手指运动时， 所述弯曲度传感器设置于五根手指的 背面； 采集肌电信号的手指数至少两根， 刺激脉冲信号可通过该肌电信号拟合再通过刺激 脉冲生成算法生成。
[0011] 优选的： 所述刺激脉冲生成算法集成在 MSP430系列微处理单片机上。
[0012] 本发明的一种基于肌电信号和传感器信号实现精细实时运动的控制方法，相比于现 有技术，具有以下有益效果： 由于采用弯曲度传感器采集主控者各肢体动作的弯曲度信息， 同时通过体表探测电极采集此刻主控者肢体动作致使相关肌肉上产生的对应肌电信号， 将 肌电信号通过刺激脉冲生成算法生成刺激脉冲信号； 将弯曲度信息作为该肌肉的刺激脉冲 信号的通道触发信号； 然后被控者通过肌电刺激电极， 对相应的肌肉施加获得的刺激脉冲 信号， 以促使被控者肢体按照主控者肢体的弯曲度信息动作， 实现精细到手指的实时动作 控制， 达到动作的准确复制和重建， 因此本发明加入传感器信号作为辅助， 来提高识别精 细动作的准确性， 尤其是手部的精细动作。 以肌电信号为主， 传感器信号为辅， 对两者综 合操作， 使用采集到的肌电信号通过刺激脉冲生成算法生成相关的刺激脉冲序列， 使得输 出刺激脉冲与原始肌电信号相关； 同时使用传感器信号作为通道触发信号， 采集肢体动作 弯曲度信息判定哪些肢体发生运动， 来选择相对应的输出刺激肢体， 将刺激脉冲输出至被 控者相对应的肌肉位点来对不同的肌肉神经进行与主控者动作强度对应的刺激， 实现精细 实时控制。 另外这样做的好处在于解决了单靠肌电信号无法识别精细动作的缺点， 同时系 统中的肌电信号弥补了传感器信号无法采集运动强度的问题， 两者信息融合可以成功地完 成动作的发生和强度记录， 将一个完整动作的要素记录还原， 并可以将其复制出来， 从而 达到精细、 实时运动控制的目的。
附图说明
[0013] 图 1是实现基于肌电信号和传感器信号实现精细到手指的实时运动控制方法的系统 示意图。
[0014] 图 2是弯曲度传感器在手上的分布示意图。
[0015] 图 3是弯曲度传感器采集信号的电路图。
[0016] 其中 1代表手指， 2代表弯曲度传感器。
具体实施方式
[0017] 附图非限制性地公开了本发明一个优选实施例的结构示意图， 以下将结合附图详细 地说明本发明的技术方案。
实施例
[0018] 本实施例的一种基于肌电信号和传感器信号实现精细实时运动的控制方法， 如图 1-3所示， 包括以下步骤： 第一步， 采用弯曲度传感器采集主控者各肢体动作的弯曲度信息， 同时通过体表探测电极 采集此刻主控者肢体动作致使相关肌肉上产生的对应肌电信号；
第二步， 通过功能电激励系统， 该功能电激励系统集成有刺激脉冲生成算法， 所述刺激脉 冲生成算法将步骤一所述弯曲度信息对应的肌电信号生成刺激脉冲信号， 所述功能电激励 系统通过步骤一所述的弯曲度信息作为对应肢体动作的刺激脉冲信号的通道触发信号； 所 述肌电信号是通过放大、 滤波、 A/D转换电路处理过； 所述刺激脉冲生成算法是指将肌电 信号生成幅度、 频率都与采集的肌电信号成正比的刺激信号。
[0019] 将弯曲度信息作为该肌肉的刺激信号的通道触发信号， 所述弯曲度信息是指通过 A/D转换电路处理过的弯曲度信息。 该通道触发信号用于选择刺激信号相应的输出刺激通 道； 所述通道触发信号的选取方法为： 首先设定一个阈值， 当弯曲度传感器收集信号低于 阈值时， 就认为发生了弯曲， 进而在对相应的输出通道选择进行脉冲刺激。
[0020] 第三步，被控者通过肌电刺激电极，对相应的肌肉施加步骤二获得的刺激脉冲信号， 以促使被控者肢体按照步骤一所述主控者肢体的弯曲度信息动作。
[0021] 当主控者所做的动作为手指运动时， 本发明是这样做的： 将弯曲度传感器设置于五 根手指的背面； 将体表探测电极分别置在手指的各个肌肉探测点上； 将肌电刺激电极置在 被控者手指的各个肌肉刺激点上； 由于有五根手指， 因此可将肌电信号分成五个通道， 该 五个通道分别对应五根手指 （拇指、 食指、 中指、 无名指和小指）， 当主控者手指运动时， 通过第一步， 即弯曲度传感器采集主控者五根手指动作的弯曲度信息， 同时通过体表探测 电极采集此刻主控者五根手指动作致使各手指肌肉上产生的对应肌电信号； 然后经过第二 步， 通过功能电激励系统， 将步骤一所述各个手指的弯曲度信息对应的肌电信号生成刺激 脉冲信号， 所述功能电激励系统通过步骤一所述各个手指的弯曲度信息作为对应手指动作 的刺激脉冲信号的通道触发信号， 该通道触发信号的选取方法为： 首先设定一个阈值， 当 手指上的弯曲度传感器收集信号低于阈值时， 就认为发生了弯曲， 进而在对相应的手指输 出通道选择进行脉冲刺激； 最后在经过第三步， 被控者通过肌电刺激电极， 对相应的手指 肌肉施加步骤二获得的刺激脉冲信号， 以促使被控者手指按照步骤一所述主控者手指的弯 曲度信息动作。 比如若主控者做了弯曲食指的动作， 这时食指上的弯曲度传感器的阻抗发 生改变， 输出电压值改变， 经过 AD后送入功能电激励系统， 功能电激励系统的算法会监 测到对应于食指的传感器信号发生了变动， 而其他传感器信号都没有变动， 就可以判断出 主控者手部的动作是食指弯曲， 就可以产生让被控者食指弯曲的相应的刺激脉冲。 在手指 运动过程中， 体表探测电极可以只测量两根手指动作生成的肌电信号， 即只测量两通道的 肌电信号， 而刺激脉冲信号可通过该肌电信号拟合再通过刺激脉冲生成算法生成， 而手指 的动作可通过弯曲度传感器判定。
[0022] 本发明的原理： 一种基于肌电信号和传感器信号实现精细到手指的实时运动控制方 法， 其核心思想在于： 采集主控者动作相关肌肉上的肌电信号， 通过刺激脉冲生成算法生 成相关的刺激脉冲序列， 再利用弯曲度传感器采集到的各个手指弯曲度信息作为通道触发 信号， 选择相应的输出刺激通道， 将生成的刺激脉冲序列输出至被控者相关的肌肉刺激位 点上， 从而实现对不同的目标肌肉神经进行与源神经 /肌电信号相关的功能电刺激， 实现精 细到手指的精细动作实时控制。
[0023] 传感器信号由弯曲度传感器采集而来。弯曲传感器的原理是弯曲程度的改变导致自 身阻抗的改变， 进而达到感知弯曲程度的目的。 通过这些信号建立还原算法， 可以通过精 确测量每一根手指的弯曲度和手指之间的相对运动， 还原手部的动作。
[0024] 传感器信号的引入主要是为了解决精细动作的识别问题， 单单依靠肌电信号可以识 别一些粗大动作如抓握， 但是对于手指的运动， 单靠肌电信号很难识别， 引入弯曲度传感 器可以监测每一根手指的运动弯曲状况， 进而达到识别精细动作的目的。
[0025] 肌电信号是通过放大、 滤波、 A/D转换电路采集的人体特定部位的肌电信号。 采集 后送入信号处理电路， 一般根据相应动作的复杂程度来决定选取多少路的肌电信号， 本发 明选取两路肌电信号再配合传感器信号就足以达到很好的效果。
[0026] 将弯曲度传感器信号作为开关信号， 具体做法是设定一个阈值， 当弯曲度传感器收 集信号低于阈值时， 就认为手指发生了弯曲， 进而在对相应的输出通道进行脉冲刺激， 达 到主控者受控者同步的目的。
[0027] 采集特定肌肉上的肌电信号通过刺激脉冲生成算法生成相关的刺激脉冲序列， 具体 是指通过采集主控者特定肌肉上的肌电信号， 利用算法生成幅度、 频率都与采集到的肌电 信号成正比的刺激脉冲序列， 使得输出信号与采集到的主控者的肌电信号相关且成正比， 当主控者动作幅度大时， 产生的肌电信号较大， 输出刺激脉冲也相应较大， 使得主控者和 受控者之间有良好的反馈关联。
[0028] 精确到手指运动的精细动作控制是指将利用传感器信号和肌电信号产生的多路刺 激脉冲来刺激被控者的特定肌肉位点， 就可以让相应对象进行精确的运动， 这种运动的精 细程度可以达到单根手指运动的级别。
[0029] 实时是指在采集者做出动作后， l-2s 内受刺激对象就可以做出相应的精细动作响 应。 信号采集处理以及算法是一个非常快的过程， 其耗时在毫秒级别， 主要的延时产生于 肌肉对于刺激脉冲的响应时间， 因为信号要经过一定时间才能产生让肌肉运动的能量， 尤 其在比较微弱的动作情况下， 这种延时会更为明显。 幅度较大的动作如抓握则延时较短， 刺激对象的动作与采集对象的动作是几乎同步的。
[0030] 刺激脉冲生成算法是通过单片微型计算机实现的， 在实验中， 采用的是 TI公司的 MSP430 系列微处理器， 采用其他系列的微处理器可以同样实现相同的算法。 在算法处理 后， 将生成的刺激脉冲信号输入到刺激电路中， 由刺激电路产生最终的刺激脉冲波形， 并 通过体表电极将刺激脉冲输出至人体相应运动肌肉的敏感控制位点。
[0031] 基于肌电信号和传感器实现精细到手指运动控制的方法可以实现健康人带动瘫痪 病人实时的做精细动作， 或者瘫痪病人健康肢体带动瘫痪肢体进行实时的相同的精细动 作， 或者健康人带动一个或者多个健康人进行实时的精细动作， 实现动作的准确复制和重 建， 无论是对瘫痪病人的康复还是健康人的操作训练， 这些功能都可以产生积极的帮助 本发明的特点在于： 引入肌电信号和传感器信号相互融合的理念， 采集肌电信号生成相应 的刺激脉冲序列， 加入传感器信号作为开关信号来选择输出通道， 通过两者的融合， 可以 识别动作的发生和强度， 并将结果输出至特定的受刺激肌肉， 使之产生相同的运动。 这样 做的好处在于解决了单靠肌电信号无法识别精细动作的缺点， 同时系统中的肌电信号弥补 了传感器信号无法采集运动强度的问题， 两者信息融合可以成功地完成动作的发生和强度 记录， 将一个完整动作的要素记录还原， 并可以将其复制出来， 从而达到精细到手指的实 时运动控制。 用这种方法可以让健康人带动瘫痪病人做实时精细动作， 或者瘫痪病人健康 肢体带动瘫痪肢体进行实时的相同的精细动作， 或者健康人带动一个或者多个健康人进行 精细动作，实现动作的准确复制和重建，无论是对瘫痪病人的康复还是健康人的操作训练， 这些功能都可以有积极的帮助。
[0032] 本发明的具体实施案例，利用该方法实现健康人带动瘫痪病人进行实时精细运动的 实例。 首先， 利用体表电极采集主控者特定肌肉的肌电信号， 通过 AD转化后变为数字信 号。 同时， 利用弯曲度传感器信号采集主控者手指的运动信息， 其弯曲度传感器在手上的 分布如图 2所示， 其采集信号的电路图如图 3所示。 通过这种分布， 可以监测每一根手指 的弯曲度信息， 进而还原出手部的精确动作。 根据传感器特性， 用图 4中的电路来将传感 器采集的模拟信号转化为数字信号， 其中 Rf 代表弯曲度传感器， 传感器会随着弯曲度的 变化而改变自身阻抗， 进而改变输出电压， 在经过 AD转换后就可以将弯曲度的信息用电 压来表现。
[0033] 然后， 将通过 AD转化肌电和传感器两种数字信号送入 MCU中。 样机中 MCU采 用的 TI公司出品的 MSP430F169单片机。 在单片机上用软件编写相应的算法， 主要是利 用传感器采集到的弯曲度信息来判断当前的手部姿势， 每一根手指的弯曲程度， 通过前后 弯曲程度的改变来判断是否有动作发生， 构成输出刺激脉冲的选通条件 （触发条件）。 当 有特定动作发生时， 才会有相应的刺激脉冲发生， 比如主控者做了弯曲食指的动作， 这时 传感器的阻抗发生改变， 输出电压值改变， 经过 AD后送入 MCU中， MCU中的算法会监 测到对应于食指的传感器信号发生了变动， 而其他传感器信号都没有变动， 就可以判断出 主控者手部的动作是食指弯曲， 就可以产生让被控者食指弯曲的相应的刺激脉冲。 同时， MCU中的脉冲刺激生成算法根据采集到的肌电信号大小来产生相应强度的刺激脉冲序列： 当肌电信号强时， 就产生强的刺激脉冲， 当肌电信号弱时， 则产生弱的刺激脉冲。 通过肌 电信号和传感器信号的融合处理， 可以很好地实现主控者和被控者动作的同步， 使系统既 可以判断刺激种类， 又可以判断刺激强度， 达到实时精细动作控制的目的。
[0034] 刺激电路主要是根据前面算法中的关键指标来决定刺激脉冲的幅值大小和脉宽长 短， 通过刺激人体的对应位点可以达到让人做出特定动作的目的。
[0035] 通过上述步骤， 主控者做出动作， 在很短的时间内， 受控者就会做出相同的动作， 这对不论是瘫痪病人的康复还是健康人的操作训练， 都可以有积极的帮助。
[0036] 为了克服依靠纯肌电信号来识别动作的困难， 在此加入传感器信号作为辅助， 来提 高识别精细动作的准确性， 尤其是手部的精细动作。 在这个改进版的系统中， 以肌电信号 为主， 传感器信号为辅， 对两者进行一个 "与"算法操作， 进而通过算法来对不同的肌肉 神经进行与控制强度对应的刺激， 实现精细到手指运动的实时控制。
[0037] 除上述针对瘫痪病人康复医学的应用， 在体育运动员的训练， 钢琴等乐器的练习和 各种器械的操作训练等过程中， 动作的准确复制和重建也有应用价值。 设想一下， 如果能 够摆脱难以理解的口头传授的动作要领， 精确的做出教练和老师要求的标准技术动作， 无 疑节省了大量的训练时间， 更快地培养出优秀的运动员、 演奏家、 技术人员和熟练工人。
[0038] 上面结合附图所描述的本发明优选具体实施例仅用于说明本发明的实施方式， 而不 是作为对前述发明目的和所附权利要求内容和范围的限制， 凡是依据本发明的技术实质对 以上实施例所做的任何简单修改、 等同变化与修饰， 均仍属本发明技术和权利保护范畴。

Claims

权利要求书
1. 一种基于肌电信号和传感器信号实现精细实时运动的控制方法， 其特征 在于， 包括以下步骤： 步骤一， 采用弯曲度传感器采集主控者各肢体动作的弯 曲度信息，同时通过体表探测电极采集此刻主控者肢体动作致使相关肌肉上产生 的对应肌电信号； 步骤二， 通过功能电激励系统， 该功能电激励系统集成有刺激 脉冲生成算法，所述刺激脉冲生成算法将步骤一所述弯曲度信息对应的肌电信号 生成刺激脉冲信号，所述功能电激励系统通过步骤一所述的弯曲度信息作为对应 肢体动作的刺激脉冲信号的通道触发信号； 步骤三、 被控者通过肌电刺激电极， 对相应的肌肉施加步骤二获得的刺激脉冲信号，以促使被控者肢体按照步骤一所 述主控者肢体的弯曲度信息动作。
2. 根据权利要求 1 所述基于肌电信号和传感器信号实现精细实时运动的控 制方法， 其特征在于： 所述刺激脉冲生成算法是指将肌电信号生成幅度、频率都 与采集的电信号成正比的刺激信号的算法。
3. 根据权利要求 2所述基于肌电信号和传感器信号实现精细实时运动的控 制方法， 其特征在于： 所述肌电信号是指通过放大、 滤波、 A/D转换电路处理过 的肌电信号； 所述弯曲度信息是指通过 A/D转换电路处理过的弯曲度信息。
4. 根据权利要求 3所述基于肌电信号和传感器信号实现精细实时运动的控 制方法， 其特征在于： 所述通道触发信号的判断方法为： 首先设定一个阈值， 当 弯曲度传感器收集信号低于阈值时，就认为发生了弯曲， 进而在对相应的输出通 道选择进行脉冲刺激。
5. 根据权利要求 4所述基于肌电信号和传感器信号实现精细实时运动的控 制方法， 其特征在于： 当主控者所做的动作为手指运动时， 所述弯曲度传感器设 置于五根手指的背面； 采集肌电信号的手指数至少两根， 刺激脉冲信号可通过该 肌电信号拟合再通过刺激脉冲生成算法生成。
6. 根据权利要求 5所述基于肌电信号和传感器信号实现精细实时运动的控 制方法，其特征在于： 所述刺激脉冲生成算法集成在 MSP430系列微处理单片机 上。
PCT/CN2013/087570 2013-06-04 2013-11-21 一种基于肌电信号和传感器信号实现精细实时运动的控制方法 WO2014194609A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310222561.1A CN103495260B (zh) 2013-06-04 2013-06-04 一种基于肌电信号和传感器信号实现精细实时运动的控制方法
CN201310222561.1 2013-06-04

Publications (1)

Publication Number Publication Date
WO2014194609A1 true WO2014194609A1 (zh) 2014-12-11

Family

ID=49860573

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/087570 WO2014194609A1 (zh) 2013-06-04 2013-11-21 一种基于肌电信号和传感器信号实现精细实时运动的控制方法

Country Status (2)

Country Link
CN (1) CN103495260B (zh)
WO (1) WO2014194609A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109814716A (zh) * 2019-01-29 2019-05-28 福州大学 一种基于动态表面肌电信号的运动意图解码方法

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105326500B (zh) * 2014-08-13 2018-02-09 华为技术有限公司 基于表面肌电信号的动作识别方法和设备
CN104383637B (zh) * 2014-12-09 2015-12-02 北京银河润泰科技有限公司 一种培训辅助设备和培训辅助方法
CN106155300A (zh) * 2015-04-23 2016-11-23 宁波市美灵思医疗科技有限公司 一种基于肌电流和多传感器协同作用的人机交互设备及其使用方法
CN105311749B (zh) * 2015-05-11 2018-03-09 南京神桥医疗器械有限公司 一种植入式电刺激肢体运动功能的调控装置
CN105056396A (zh) * 2015-09-09 2015-11-18 上海诺诚电气股份有限公司 镜像评估训练低频电刺激康复系统及使用方法
CN107157197A (zh) * 2017-03-27 2017-09-15 韶关学院 一种基于物联网的婴儿摇篮防盗系统
CN107307681B (zh) * 2017-03-27 2023-12-29 韶关学院 一种基于物联网的婴儿摇篮智能检测系统
CN109077726A (zh) * 2018-10-29 2018-12-25 博睿康科技（常州）股份有限公司 一种肌电检测装置与肌电识别方法
CN112589775A (zh) * 2020-12-14 2021-04-02 东南数字经济发展研究院 一种外骨骼电机系统运动的控制方法
CN114706473A (zh) * 2022-03-11 2022-07-05 青岛虚拟现实研究院有限公司 一种基于学习设备的人机交互方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101732796A (zh) * 2009-12-28 2010-06-16 上海交通大学 肌电信号控制的主从式无线功能性电刺激康复系统
CN201642503U (zh) * 2010-03-11 2010-11-24 南方医科大学 基于生物反馈的监测治疗装置
CN102949783A (zh) * 2011-08-26 2013-03-06 云林科技大学 回馈控制的穿戴式上肢电刺激装置

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101391129B (zh) * 2008-07-21 2012-01-18 天津大学 基于p300信号脑机接口智能化上肢康复训练器及信号处理方法
CN105477782B (zh) * 2010-03-22 2019-01-04 纽约城市大学研究基金会 电荷增强神经电刺激系统
CN201899776U (zh) * 2010-11-15 2011-07-20 王蓉 一种智能足下垂矫正和偏瘫康复装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101732796A (zh) * 2009-12-28 2010-06-16 上海交通大学 肌电信号控制的主从式无线功能性电刺激康复系统
CN201642503U (zh) * 2010-03-11 2010-11-24 南方医科大学 基于生物反馈的监测治疗装置
CN102949783A (zh) * 2011-08-26 2013-03-06 云林科技大学 回馈控制的穿戴式上肢电刺激装置

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109814716A (zh) * 2019-01-29 2019-05-28 福州大学 一种基于动态表面肌电信号的运动意图解码方法
CN109814716B (zh) * 2019-01-29 2021-07-27 福州大学 一种基于动态表面肌电信号的运动意图解码方法

Also Published As

Publication number Publication date
CN103495260A (zh) 2014-01-08
CN103495260B (zh) 2016-05-25

Similar Documents

Publication Publication Date Title
WO2014194609A1 (zh) 一种基于肌电信号和传感器信号实现精细实时运动的控制方法
CN102727361B (zh) 坐卧式下肢康复机器人
CN103212156B (zh) 基于本体镜像原理的偏瘫患者自主康复仪
WO2020118797A1 (zh) 假肢控制方法、装置、系统、设备和存储介质
CN106236503B (zh) 肌电驱动的上肢可穿戴外骨骼系统及控制方法
WO2016091130A1 (zh) 一种培训辅助设备和培训辅助方法
CN111150935B (zh) 一种肌电颈部按摩装置以及控制方法
CN104337666A (zh) 多肌肉协同肌电反馈康复训练系统和方法
CN104107134A (zh) 基于肌电反馈的上肢训练方法及系统
WO2013029196A1 (zh) 回馈控制的穿戴式上肢电刺激装置
Chen et al. Effect of upper-limb positions on motion pattern recognition using electromyography
CN110354387B (zh) 多触发方式的智能电刺激手部训练器及方法
CN110420383A (zh) 一种基于多模态融合反馈的可调功能性电刺激控制方法
Tryon et al. Classification of task weight during dynamic motion using EEG–EMG fusion
CN109276408A (zh) 上肢手外骨骼康复机器人
CN103300853A (zh) 一种基于表面肌电的诊疗系统
Ai et al. Advanced rehabilitative technology: neural interfaces and devices
CN108543216A (zh) 一种基于主从控制的手部功能重建装置及其实现方法
Geng et al. Pattern recognition based forearm motion classification for patients with chronic hemiparesis
Chen et al. A real-time EMG-controlled functional electrical stimulation system for mirror therapy
Asres et al. A combination of AR and neural network technique for EMG pattern identification
Lou et al. Wireless master-slave FES rehabilitation system using sEMG control
CN208943270U (zh) 一种可调节的电刺激康复装置
CN215017698U (zh) 康复训练运动模拟可视化系统
Liu et al. Performance of various EMG features in identifying arm movements for control of multifunctional prostheses

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13886624

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 13886624

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 13886624

Country of ref document: EP

Kind code of ref document: A1