CN105361880B - The identifying system and its method of muscular movement event - Google Patents
The identifying system and its method of muscular movement event Download PDFInfo
- Publication number
- CN105361880B CN105361880B CN201510854222.4A CN201510854222A CN105361880B CN 105361880 B CN105361880 B CN 105361880B CN 201510854222 A CN201510854222 A CN 201510854222A CN 105361880 B CN105361880 B CN 105361880B
- Authority
- CN
- China
- Prior art keywords
- signal
- window
- event
- mean square
- myoelectricity
- Prior art date
- Legal status (The legal status 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 status listed.)
- Expired - Fee Related
Links
Landscapes
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
A kind of identifying system and its method of muscular movement event, the system include：Signal acquisition module, signal processing module and the signal identification module being made of myoelectricity acquisition module and brain wave acquisition module, wherein：Signal processing module carries out signal processing to the data that signal acquisition module acquires and carries out incident detection, and the event detected is identified signal identification module and labeled bracketing.The present invention can be detected in nervous physiology goes out effective electromyography signal with diagnostic field highly efficient labeling, that is electromyography signal event, and the myoelectricity to being marked and EEG signals event are accurately analyzed and processed, and arithmetic speed greatly improved, reduce operation delay and system equipment power consumption.
Description
Technical field
The present invention relates to a kind of technology of medical science, the identifying system of specifically a kind of muscular movement event
And its method.
Background technology
The detection of the nervous physiology of motion control and motor disorder acquires muscle activity letter simultaneously with main at present use of diagnosis
Number (such as surface electromyography) and cerebration signal (such as electroencephalogram), and cerebration and flesh are analyzed according to corresponding sports event simultaneously
The method of meat activity association.However, the selected and differentiation of corresponding sports event is relied primarily on and is accomplished manually at present, need rich
The doctor of experience handmarking in largely record data goes out the start/stop time of corresponding sports time, can realize further
Cerebration and muscle activity correlation analysis.In addition, the acquisition of muscle activity signal and cerebration signal generally requires at present
The medical institutions that patient comes profession are acquired, and the installation and use of collecting device are sufficiently complex, need the doctor of profession
Shi Jinhang operate, patient can not outside medical institutions Self-operating.So that the acquisition of muscle and cerebration signal is often
Cause time and burden economically to patient, cause larger work load to doctor, at the same also to carry out signal detection and
The doctor of diagnosis causes larger work load.
By the retrieval discovery to the prior art, Chinese patent literature CN101057795, open (bulletin) day
2007.10.24, disclose it is a kind of using myoelectricity and the prosthetic hand and its control method of brain electricity Collaborative Control, using myoelectricity and brain
The prosthetic hand of electric Collaborative Control, myoelectricity brain electricity power-collecting electrode, myoelectricity EEG Processing module, the acquisition of A/D change datas, myoelectricity
Signal movement pattern-recognition and trajectory prediction module, electrical artificial limb hand, tactile and slip sense integrated transducer, system feedback stimulation dress
It puts, strength and speed adjustment module.Using the method for myoelectricity and brain electricity Collaborative Control prosthetic hand, include the following steps：Myoelectricity is believed
It number is acquired and amplifies；Feature extraction and pattern-recognition；The crawl situation of object is grabbed in detection；It is not captured when being grabbed object
It is good, to the physical stimulation signal of manipulator's certain forms；Detect brain electric information；It is output to strength and speed adjustment module；Output
Signal completes the further control to electric hand.The feature extraction and pattern-recognition of the technology need continuous real-time window signal
Processing, i.e., while signal acquisition, the continuous signal numerical value intercepted in certain time window, and corresponding feature is applied to it
Extraction and pattern-recognition.This mode can ensure the real-time processing to effective electromyography signal, but need to consume a large amount of operations, wave
Take the invalid electromyography signal of a large amount of calculation process, often result in the problem of operand is excessive, operation delay is high.In addition, it is controlled in movement
The detection of the nervous physiology of system and motor disorder and diagnostic field, doctor do not need to real-time electromyography signal and handle, but it is long when
The electromyographic signal collection of journey, and efficiently find out effective electromyography signal and processing analysis is carried out to it.Therefore the technology is applied to god
Excessive invalid operation can be caused through physiological detection and diagnostic field, operand is excessive, and operation delay is high, and system power dissipation is high to ask
Topic.
Invention content
The present invention proposes identifying system and its side of a kind of muscular movement event for deficiencies of the prior art
Method can detect in nervous physiology and go out effective electromyography signal, i.e. electromyography signal event, and to being marked with diagnostic field highly efficient labeling
The myoelectricity and EEG signals event of note are accurately analyzed and processed, and arithmetic speed greatly improved, reduce operation delay and
System equipment power consumption.
The present invention is achieved by the following technical solutions：
The present invention relates to a kind of identifying system of muscular movement event, including：By myoelectricity acquisition module and brain wave acquisition mould
Signal acquisition module, signal processing module and the signal identification module of block composition, wherein：Signal processing module is to signal acquisition mould
The data of block acquisition carry out signal processing and carry out incident detection, and signal identification module is identified and marks to the event detected
It scores class.
The myoelectricity acquisition module is made of the wireless myoelectric sensor of single or multiple small volumes, can be attached to human body
Surface skin at collection surface electromyography signal, and institute's gathered data is wirelessly transmitted to signal processing module.
The brain wave acquisition module is made of more the single wireless brain wave acquisition equipment led, brain wave acquisition equipment that this is wireless
For 16 to 64 channel standard brain wave acquisition equipment, and pass through built-in wireless signal transmission equipment and emit collected EEG signals
Or receive control instruction.
The signal processing module includes：Average reference unit, adaptive-filtering unit, window events probe unit altogether
With event triggered mark unit and signal cutting unit, wherein：Average reference unit receives the brain electricity mould of brain wave acquisition module altogether
Intend signal and export effective brain electric information to signal cutting unit, adaptive-filtering unit is connected and connects with myoelectricity acquisition module
Adductor muscle electric analoging signal carries out time sequence window label by window events probe unit after filtered processing to myoelectricity digital signal,
Event triggered mark unit carries out the detection of rising edge or failing edge from the myoelectricity digital signal after label, and myoelectricity is effective
Event information is exported to signal cutting unit, and signal cutting unit is according to the triggered time to from the effective of common average reference unit
Brain electric information and myoelectricity validity event information are split and carry out dimension-reduction treatment, and export to signal identification module.
The dimension-reduction treatment refers to：It reduces the sample rate of signal and multi channel signals is spliced into one-dimensional signal.
The adaptive-filtering unit includes：Spectra calculation component, noise trap component and bandpass filtering component,
In：Spectra calculation module carries out spectra calculation after myoelectricity analog signal is carried out analog-to-digital conversion, and composes threshold with built-in power
Value is compared, and the brain electricity digital signal being calculated and myoelectricity digital signal are exported to noise trap component, noise trap
Component be filtered filter make an uproar handle and after obtained denoising information is filtered by bandpass filtering component output to window events
Probe unit.
The window events probe unit includes：Window moving assembly, root mean square calculation component and Threshold Detection component,
Wherein：Window moving assembly receives the denoising information after bandpass filtering, and carries out cutting processing using rectangular window, will believe after cutting
It number is sequentially output to root mean square calculation component and carries out root mean square calculation, Threshold Detection component is to the power signal after root mean square calculation
Threshold decision is carried out, and root mean square/performance number of either window signal is more than signal root mean square/power in the window and is averaged
The window signal of the preset multiple of value is exported to event triggered mark unit.
Window signal after cutting is carried out root mean square calculation by the root mean square calculation component, and judges the window signal
In latter half root mean square/power whether be more than first half root mean square/power root mean square multiple, when more than when should
Window signal is exported to Threshold Detection component, otherwise gives up the window signal.
The signal identification module includes：Pattern recognition unit and event flag feedback unit, wherein：Pattern-recognition list
Member carries out pattern-recognition according to the brain electric analoging signal after segmentation and obtains classification information, and event flag feedback unit provides man-machine
Interactive interface carries out verification and update mark result and classification information manually for user, and corresponding manual modification record will be saved into
In the model library of pattern recognition unit, algorithm for pattern recognition is enable constantly to store data and improves identification accuracy.
The pattern-recognition believes event using supervised learning algorithm, semi-supervised learning algorithm and unsupervised learning algorithm
Number carry out pattern-recognition and classification.
Classification in the classification information includes but not limited to：Proper motion, non-autonomous movement, movement of trembling, psychology
Property motor disorder, myasthenia type motor disorder and nerve motor disorder.
Technique effect
Compared with prior art, the present invention realizes that system operations rate promotes more than 50% compared with the prior art, system power dissipation
33% is reduced compared with the prior art so that doctor is carrying out related neural physiological detection with that can greatly improve working efficiency during diagnosis,
Reduce the signal processing and analyzing time.
The present invention also can directly be run on wearable hardware, reduce operation consumption, make to carry out with wearable hardware
Long time-histories is recorded and is analyzed feasible.The prior art is difficult to run directly on wearable hardware since power consumption is excessive so that doctor
All kinds of electromyography signal events and EEG signals event can be marked, classified and handled with more convenient.It is efficiently quick to establish
Corresponding mathematical statistical model.
Description of the drawings
Fig. 1 is present invention processing schematic diagram；
Fig. 2 is the method for the present invention schematic diagram；
Fig. 3 handles schematic diagram for adaptive-filtering；
Fig. 4 is window events detection processing schematic diagram.
Specific embodiment
As depicted in figs. 1 and 2, the present embodiment is related to the recognition methods of above system, specifically includes following steps：
Step 1, acquisition muscle activity analog signal and brain activity analog signal；
The present embodiment by signal acquisition module realizes and acquires that the signal acquisition module is adopted by myoelectricity acquisition module and brain electricity
Collect module composition.
The myoelectricity acquisition module is made of the wireless myoelectric sensor of single or multiple small volumes, number of sensors
It is 1~20, preferably 4, can be attached at the surface skin of human body, collection surface electromyography signal, by the way of wireless telecommunications
Institute's gathered data is transmitted to signal processing module.
In the present embodiment in a manner that bluetooth 4.0 is as the wireless telecommunications.
The brain wave acquisition module is made of more the single wireless brain wave acquisition equipment led, brain wave acquisition equipment that this is wireless
Port number be 4~64, preferably 32.
Step 2 handles the signal of acquisition and carries out incident detection, specially：To myoelectricity analog signal successively into
Row adaptive-filtering, window events detection and the processing of event triggering moment label, obtain myoelectricity validity event information；Simultaneously to brain
Electric analoging signal carries out common average reference processing, obtains effective brain electric information；Finally according to the time of myoelectricity validity event and institute
If window width obtains event set to be identified.
As shown in figure 3, the adaptive-filtering, includes the following steps：
Step 1 carries out Signal Pretreatment respectively to electromyography signal and EEG signals, specially：Using common average reference
(CAR), Kalman filtering, dynamic window averaging method filter off the noise signal in all channels.
Step 2, power spectrum signal are calculated, are compared with predetermined power spectrum, specially：Calculate power spectrum signal, by result with
Predetermined power modal data group carries out pattern-recognition, and mode identification method is approximate, refreshing using KNN, gauss hybrid models, maximum likelihood
Through network.
Step 3, noise trap and electromyography signal bandpass filtering obtain effective electromyography signal, specifically from acquisition signal
For：Invalid signals part is filtered off using high-pass filtering, low-pass filtering, band-pass filtering method, retains useful signal.It is very filtered using promise
Wave method trap removes Hz noise and characteristic frequency.
As shown in figure 4, the window events detection, includes the following steps：
Step 1, partition window, specially：Signal is divided into small fragment one by one and carries out incident detection respectively, it is specific to walk
Suddenly it is：Window width, stride setting, window width are 0.1s~60s, and preferably 5s, stride is 0.1s~60s, preferably 1s.
Step 2, setting window threshold value simultaneously judge：The condition of window reception threshold determination is the window in the present embodiment
Root mean square/power of the latter half of signal is more than the certain multiple of first half root mean square/power in mouthful, multiple for 2~
100, preferably 5 times meet decision condition and then carry out further incident detection to the window, otherwise skip and change window judgement
Next window；
Step 3, Threshold Detection：By root mean square/power of signal in calculation window, can further be calculated in window
Average root-mean-square/power of signal, the decision condition of Threshold Detection is root mean square/performance number of certain time-ofday signals in the present embodiment
More than the certain multiple of the average value of signal root mean square/power in the window, multiple is 1~100 times, preferably 5 times.Described in satisfaction
Threshold Detection decision condition at the time of i.e. be marked as event triggering at the time of.
Preferably, window events detection, time are no longer carried out in the intervals after marked event time
Between be divided into 0.1~60s, preferably 1.5s.It can ensure that window events detection will not be to the event signal among same event in this way
It is marked, so as to improve the accuracy of event flag.
The common average reference processing, specially：
Step 1, the EEG signals for choosing a channel, calculate the average value of the channel signal, and subtract with the average value
The signal value of all signals；
Step 2 obtains the opposite signal value with the average value of all signals, can be greatly reduced in EEG signals in this way
Noise improves the efficiency of EEG Processing analysis.
Step 3, after above-mentioned event triggering moment label is completed, according to corresponding triggering moment, to EEG signals and flesh
Electric signal carries out signal segmentation, and the signal of segmentation number is stored as to the signal of corresponding event respectively.
In step 3 preferably：One segmentation section of setting, the beginning and ending time is respectively 0.1~60s and thing before event triggering moment
0.1~60s after part triggering moment, preferably beginning and ending time are 3s after 3s to event triggering moment before event triggering moment.
Step 3 carries out pattern-recognition and labeled bracketing to the event detected.
The present embodiment specific implementation records data, data based on 15 nervous physiologies that Ruishiwo State Central Hospital provides altogether
Form is 32 channel brain electricity and 9 channel myoelectricities, is recorded 1 hour often.Wherein, 5 number of cases are according to the autonomous fortune having an effect for normal muscle
Dynamic data, 10 non-autonomous exercise datas having an effect for ill muscle.In 10 non-autonomous exercise datas, 5 are amyostasia
Case, 5 be Parkinson's case.
The method for event of being had an effect using traditional handmarking's muscle, specialist analysis is per an example record number of 1 hour
According to about 1 hour event is needed, event of having an effect each time in data is marked and analyzed.Completing 15 data analyses then needs
15 hours.
Using this method, 15 number of cases according to can complete whole event flags and analysis in 5 minutes, and can significant notation go out
Autokinetic movement time or non-autonomous motion event, rate of accuracy reached 98.5%.To different case events also can accurate marker, accurately
Rate is up to 83%.The corresponding EEG signals cortex action readiness potential of event of having an effect each muscle
(Bereitschaftspotential) it can accurately be recorded.It greatly facilitates neurologist and carries out Correlation method for data processing
And analysis.
Above-mentioned specific implementation can by those skilled in the art under the premise of without departing substantially from the principle of the invention and objective with difference
Mode carry out local directed complete set to it, protection scope of the present invention is subject to claims and not by above-mentioned specific implementation institute
Limit, each implementation within its scope is by the constraint of the present invention.
Claims (9)
1. a kind of identifying system of muscular movement event, including：The signal being made of myoelectricity acquisition module and brain wave acquisition module
Acquisition module, signal processing module and signal identification module, wherein：The data that signal processing module acquires signal acquisition module
It carries out signal processing and carries out incident detection, the event detected is identified signal identification module and labeled bracketing；
The signal processing module includes：Average reference unit, adaptive-filtering unit, window events probe unit and thing altogether
Part triggered mark unit and signal cutting unit, wherein：Average reference unit receives the brain electrical analogue letter of brain wave acquisition module altogether
Number and effective brain electric information is exported to signal cutting unit, adaptive-filtering unit is connected with myoelectricity acquisition module and receives flesh
Electric analoging signal carries out time sequence window label, event by window events probe unit after filtered processing to myoelectricity digital signal
Triggered mark unit carries out the detection of rising edge or failing edge from the myoelectricity digital signal after label, and by myoelectricity validity event
Information is exported to signal cutting unit, and signal cutting unit is according to the triggered time to effective brain electricity from common average reference unit
Information and myoelectricity validity event information are split and carry out dimension-reduction treatment, and export to signal identification module；
The window events probe unit includes：Window moving assembly, root mean square calculation component and Threshold Detection component,
In：Window moving assembly receives the denoising information after bandpass filtering, and carries out cutting processing using rectangular window, by signal after cutting
Be sequentially output to root mean square calculation component carry out root mean square calculation, Threshold Detection component to the power signal after root mean square calculation into
Row threshold decision, and the root mean square of either window signal or performance number are more than being averaged for signal root mean square in the window or power
The window signal of the preset multiple of value is exported to event triggered mark unit.
2. system according to claim 1, it is characterized in that, the adaptive-filtering unit includes：Spectra calculation group
Part, noise trap component and bandpass filtering component, wherein：After myoelectricity analog signal is carried out analog-to-digital conversion by spectra calculation module
Spectra calculation is carried out, and is compared with built-in power spectrum threshold value, by the brain electricity digital signal and myoelectricity that are calculated number
Signal is exported to noise trap component, and noise trap component is filtered to filter to make an uproar and handles and filter obtained denoising information by band logical
Wave component is exported after being filtered to window events probe unit.
3. system according to claim 1, it is characterized in that, the root mean square calculation component is by the window signal after cutting
Carry out root mean square calculation, and judge the root mean square of latter half in the window signal or power whether to be more than first half square
The root mean square multiple of root or power, when more than when the window signal exported to Threshold Detection component, otherwise give up window letter
Number.
4. system according to claim 1, it is characterized in that, the signal identification module includes：Pattern recognition unit and
Event flag feedback unit, wherein：Pattern recognition unit carries out pattern-recognition according to the brain electric analoging signal after segmentation and obtains
Classification information, event flag feedback unit provide human-computer interaction interface for the manual verification of user's progress and update mark result and divide
Category information, corresponding manual modification record will be saved into the model library of pattern recognition unit, enable algorithm for pattern recognition not
Disconnected storage data simultaneously improve identification accuracy.
5. system according to claim 4, it is characterized in that, the pattern-recognition is using supervised learning algorithm, semi-supervised
Learning algorithm and unsupervised learning algorithm carry out pattern-recognition and classification to event signal.
6. a kind of recognition methods of muscular movement event according to the system any in Claims 1 to 5, which is characterized in that
First by acquiring muscle activity analog signal and brain activity analog signal；Then myoelectricity analog signal is carried out successively certainly
Adaptive filtering, window events detection and the processing of event triggering moment label, obtain myoelectricity validity event information；Simultaneously to brain electricity mould
Intend signal and carry out common average reference processing, obtain effective brain electric information；Finally according to the time of myoelectricity validity event and set window
Mouth width degree obtains event set to be identified, carries out pattern-recognition to the event detected and labeled bracketing is achieved.
7. recognition methods according to claim 6, it is characterized in that, the adaptive-filtering includes the following steps：
Step 1 carries out Signal Pretreatment respectively to electromyography signal and EEG signals：Using common average reference, Kalman filtering or
Dynamic window averaging method filters off the noise signal in all channels；
Step 2, power spectrum signal are calculated, are compared with predetermined power spectrum：Power spectrum signal is calculated, result and predetermined power are composed into number
Pattern-recognition is carried out according to group, mode identification method is using KNN, gauss hybrid models, maximum likelihood approximation or neural network；
Step 3, noise trap and electromyography signal bandpass filtering obtain effective electromyography signal from acquisition signal：Using high pass
Filtering, low-pass filtering and band-pass filtering method filter off invalid signals part, retain useful signal, using the strange filtering method trap of promise
Remove Hz noise and characteristic frequency.
8. recognition methods according to claim 7, it is characterized in that, window events detection includes the following steps：
Step 1, partition window：Signal is divided into small fragment one by one and carries out incident detection respectively, the specific steps are：Window is wide
Degree and stride setting, window width are 0.1s~60s, and stride is 0.1s~60s；
Step 2, setting window threshold value simultaneously judge：The condition of window reception threshold determination is the latter half of of signal in the window
The root mean square or power that divide are more than the multiple of first half root mean square or power, and multiple is 2~100, and it is then right to meet decision condition
The window carries out further incident detection, otherwise skips and changes the next window of window judgement；
Step 3, Threshold Detection：By the root mean square or power of signal in calculation window, being averaged for signal in window is calculated
Root mean square or power, wherein：The decision condition of Threshold Detection is more than for the root mean square or performance number of certain time-ofday signals in the window
The certain multiple of the average value of signal root mean square or power, multiple is 1~100 times, at the time of meeting Threshold Detection decision condition
At the time of being marked as event triggering.
9. recognition methods according to claim 7, it is characterized in that, common average reference processing, specially：
Step 1, the EEG signals for choosing a channel, calculate the average value of the channel signal, and are subtracted with the average value all
The signal value of signal；
Step 2 obtains the opposite signal value with the average value of all signals；
Step 3, after above-mentioned event triggering moment label is completed, according to corresponding triggering moment, EEG signals and myoelectricity are believed
Number signal segmentation is carried out, the signal divided number is stored as the signal of corresponding event respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510854222.4A CN105361880B (en) | 2015-11-30 | 2015-11-30 | The identifying system and its method of muscular movement event |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510854222.4A CN105361880B (en) | 2015-11-30 | 2015-11-30 | The identifying system and its method of muscular movement event |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105361880A CN105361880A (en) | 2016-03-02 |
| CN105361880B true CN105361880B (en) | 2018-06-26 |
Family
ID=55364870
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510854222.4A Expired - Fee Related CN105361880B (en) | 2015-11-30 | 2015-11-30 | The identifying system and its method of muscular movement event |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105361880B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108209911A (en) * | 2017-05-25 | 2018-06-29 | 深圳市前海未来无限投资管理有限公司 | A kind of electromyographic signal collection method and device |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105771087A (en) * | 2016-04-06 | 2016-07-20 | 上海乃欣电子科技有限公司 | Rehabilitation training system based on music and myoelectricity feedback simulating |
| CN105879329A (en) * | 2016-04-06 | 2016-08-24 | 上海乃欣电子科技有限公司 | Respiratory training system and method based on bioelectric signals |
| CN106529476B (en) * | 2016-11-11 | 2019-09-10 | 重庆邮电大学 | A kind of EEG feature extraction and classification method stacking network based on deep layer |
| CN107361773B (en) * | 2016-11-18 | 2019-10-22 | 深圳市臻络科技有限公司 | For detecting, alleviating the device of Parkinson's abnormal gait |
| CN106805968A (en) * | 2016-12-20 | 2017-06-09 | 广州视源电子科技股份有限公司 | Electroencephalogram relaxation degree identification method and device |
| CN108209912B (en) * | 2017-05-25 | 2020-06-05 | 深圳市前海未来无限投资管理有限公司 | Electromyographic signal acquisition method and device |
| CN107862302A (en) * | 2017-11-29 | 2018-03-30 | 合肥赑歌数据科技有限公司 | A kind of human motion detecting system and method based on semi-supervised learning |
| WO2019127559A1 (en) * | 2017-12-30 | 2019-07-04 | 深圳迈瑞生物医疗电子股份有限公司 | High-frequency radio frequency interference removal apparatus and method |
| CN109460144A (en) * | 2018-09-18 | 2019-03-12 | 逻腾（杭州）科技有限公司 | A kind of brain-computer interface control system and method based on sounding neuropotential |
| CN109318207B (en) * | 2018-11-07 | 2020-11-10 | 西安交通大学 | Myoelectric timing lower limb movement preparatory potential detection system and method |
| CN109700472A (en) * | 2019-02-21 | 2019-05-03 | 北京七鑫易维信息技术有限公司 | A kind of fatigue detection method, device, equipment and storage medium |
| CN110123335B (en) * | 2019-05-21 | 2022-06-28 | 首都医科大学宣武医院 | Information processing method, device and system |
| CN111012341B (en) * | 2020-01-08 | 2022-04-22 | 东南大学 | Artifact removal and electroencephalogram signal quality evaluation method based on wearable electroencephalogram equipment |
| CN111603162B (en) * | 2020-05-07 | 2023-05-30 | 京东科技信息技术有限公司 | Myoelectric signal processing method and device, intelligent wearable equipment and storage medium |
| CN113633887A (en) * | 2021-07-26 | 2021-11-12 | 江苏福瑞科技有限公司 | Biostimulation feedback system based on filtering algorithm |
| CN113812964B (en) * | 2021-08-02 | 2023-08-04 | 杭州航弈生物科技有限责任公司 | Proxy measurement and pseudo-multimode frozen gait detection method and device for electroencephalogram characteristics |
| CN113967023B (en) * | 2021-10-27 | 2023-08-29 | 复旦大学 | Closed-loop optogenetic intervention system and intervention method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101332136A (en) * | 2008-08-01 | 2008-12-31 | 杭州电子科技大学 | Electric artificial hand combined controlled by brain electricity and muscle electricity and control method |
| CN101464728A (en) * | 2009-01-05 | 2009-06-24 | 清华大学 | Human-machine interaction method with vision movement related neural signal as carrier |
| CN101732110A (en) * | 2009-11-19 | 2010-06-16 | 杭州电子科技大学 | Method for fusing brain electricity and muscle electricity signal chaos characteristics for hand motion identification |
| CN101776981A (en) * | 2010-01-07 | 2010-07-14 | 天津大学 | Method for controlling mouse by jointing brain electricity and myoelectricity |
| CN102488515A (en) * | 2011-12-09 | 2012-06-13 | 天津大学 | Conjoint analysis method for electroencephalograph and electromyography signals based on autonomous movement and imagination movement |
-
2015
- 2015-11-30 CN CN201510854222.4A patent/CN105361880B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101332136A (en) * | 2008-08-01 | 2008-12-31 | 杭州电子科技大学 | Electric artificial hand combined controlled by brain electricity and muscle electricity and control method |
| CN101464728A (en) * | 2009-01-05 | 2009-06-24 | 清华大学 | Human-machine interaction method with vision movement related neural signal as carrier |
| CN101732110A (en) * | 2009-11-19 | 2010-06-16 | 杭州电子科技大学 | Method for fusing brain electricity and muscle electricity signal chaos characteristics for hand motion identification |
| CN101776981A (en) * | 2010-01-07 | 2010-07-14 | 天津大学 | Method for controlling mouse by jointing brain electricity and myoelectricity |
| CN102488515A (en) * | 2011-12-09 | 2012-06-13 | 天津大学 | Conjoint analysis method for electroencephalograph and electromyography signals based on autonomous movement and imagination movement |
Non-Patent Citations (1)
| Title |
|---|
| 结合脑电信息的多自由度肌电假手研究;唐建友;《CNKI 中国优秀硕士电子期刊全文数据库（电子期刊）医药卫生科技辑》;20120331;第14-17页，第21页，第48页，第53页，图5-2，图5-4 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108209911A (en) * | 2017-05-25 | 2018-06-29 | 深圳市前海未来无限投资管理有限公司 | A kind of electromyographic signal collection method and device |
| CN108209911B (en) * | 2017-05-25 | 2020-02-07 | 深圳市前海未来无限投资管理有限公司 | Electromyographic signal acquisition method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105361880A (en) | 2016-03-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105361880B (en) | The identifying system and its method of muscular movement event | |
| CN106726030B (en) | Brain machine interface system and its application based on Clinical EEG Signals control robot movement | |
| CN110070105B (en) | Electroencephalogram emotion recognition method and system based on meta-learning example rapid screening | |
| TW201918222A (en) | Method and apparatus for high accuracy photoplethysmogram based atrial fibrillation detection using wearable device | |
| JP2020536693A (en) | ECG automatic analysis method and device based on artificial intelligence self-learning | |
| CN113397555A (en) | Arrhythmia classification algorithm of C-LSTM for physiological parameter monitoring | |
| CN109758145B (en) | Automatic sleep staging method based on electroencephalogram causal relationship | |
| CN111544015B (en) | Cognitive power-based control work efficiency analysis method, device and system | |
| CN114532993B (en) | Automatic detection method for electroencephalogram high-frequency oscillation signals of epileptic patients | |
| CN104914994A (en) | Aircraft control system and fight control method based on steady-state visual evoked potential | |
| Li et al. | EEG signal classification method based on feature priority analysis and CNN | |
| CN109044280B (en) | Sleep staging method and related equipment | |
| CN103405225B (en) | A kind of pain that obtains feels the method for evaluation metrics, device and equipment | |
| CN111598451A (en) | Control work efficiency analysis method, device and system based on task execution capacity | |
| George et al. | Sensing, processing and application of EMG signals for HAL (Hybrid Assistive Limb) | |
| CN100998503A (en) | Method for automatically recogniting and eliminating ophthalmogyric interference in electroencephalo-signals | |
| CN114052744A (en) | Electrocardiosignal classification method based on pulse neural network | |
| Allam et al. | Customized deep learning algorithm for drowsiness detection using single-channel EEG signal | |
| Wu et al. | BECT spike detection algorithm based on optimal template matching and morphological feature selection | |
| CN204759349U (en) | Aircraft controlling means based on stable state vision evoked potential | |
| Rosli et al. | Electrocardiographic (ECG) and Electromyographic (EMG) signals fusion for physiological device in rehab application | |
| Chang et al. | A versatile wireless portable monitoring system for brain–behavior approaches | |
| CN110705656A (en) | Facial action recognition method based on EEG sensor | |
| CN114237383B (en) | Multi-state identification method based on forehead single-lead electroencephalogram signals | |
| CN114098768B (en) | Cross-individual surface electromyographic signal gesture recognition method based on dynamic threshold and EasyTL |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180626 Termination date: 20191130 |