CN112162634A - Digital input brain-computer interface system based on SEEG signal - Google Patents
Digital input brain-computer interface system based on SEEG signal Download PDFInfo
- Publication number
- CN112162634A CN112162634A CN202011012679.8A CN202011012679A CN112162634A CN 112162634 A CN112162634 A CN 112162634A CN 202011012679 A CN202011012679 A CN 202011012679A CN 112162634 A CN112162634 A CN 112162634A
- Authority
- CN
- China
- Prior art keywords
- module
- data
- electroencephalogram
- stimulation
- seeg
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
- G06F3/015—Input arrangements based on nervous system activity detection, e.g. brain waves [EEG] detection, electromyograms [EMG] detection, electrodermal response detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Abstract
The invention discloses a digital input brain-computer interface system based on SEEG signals, which comprises a stimulation presenting module, a signal collecting module, a model training module and an online data analyzing module, wherein the stimulation presenting module is used for presenting stimulation signals; the stimulation presenting module presents a P300 electroencephalogram mode for stimulating and inducing a testee in a single flashing mode; the signal acquisition module is used for acquiring and amplifying intracranial electroencephalogram signals of a testee; the model training module carries out classifier training on the training data to obtain a classifier response model; and the online data analysis module analyzes the online data in real time by using the obtained classifier response model and outputs a result. The system of the invention can realize the digital input function by decoding the intracranial electrode signal. The system can achieve the same or higher input accuracy and information transfer rate by using a smaller number of electrode channels. The invention brings a brand-new interactive experience for some epileptics, also provides an effective way for clinical scientific research, and has potential social value.
Description
Technical Field
The invention relates to the technical field of biomedical signal processing technology and brain-computer interfaces, in particular to a digital input brain-computer interface system based on SEEG (deep intracranial stereotactic electroencephalogram) signals.
Background
The brain-computer interface technology has become a novel man-machine interaction mode in recent years, a user can directly utilize brain signals to interact with a computer without limb movement, interaction experience of healthy people is improved, and a reliable mode of interacting with the outside is provided for patients with incomplete motion functions or other diseases. Most brain-computer interface systems currently acquire non-implantable electrophysiological signals, such as EEG, fMRI, NIRS, etc. However, these signals have limitations in terms of time and spatial resolution, and are susceptible to external interference, and the signal noise is large, which has a great influence on the performance and efficiency of the system.
SEEG is an implanted electroencephalogram acquisition technology, is widely applied to surgical treatment of refractory epilepsy in recent years, and plays a key role in diagnosis of epileptic focus and prediction of epileptic seizure. The SEEG electrodes are very thin elongated cylinders that can reach deep into the cranium, to deep brain regions and to some very shallow cortex. A plurality of electrode contacts (the number is from 8 to 20) are sequentially arranged on the electrodes at equal intervals, and the electrode contacts can record electroencephalogram signals and can also apply electrical stimulation. The SEEG electrode can be accurately placed in a deep intracranial area to acquire intracranial pure electroencephalogram signals, signal interference is small, signal-to-noise ratio is high, and other electroencephalogram acquisition technologies cannot achieve the signal interference, so the SEEG electrode is an effective means for constructing a high-performance brain-computer interface.
P300 is a most highly interesting event-related potential (ERP), a special evoked potential. When a subject is subjected to an OB stimulation sequence (oddball paradigm) of visual, auditory or somatosensory, a positive potential, known as the P300 potential, is generated in the brain electrical signal about 300 milliseconds after stimulation. The brain-computer interface constructed based on the P300 potential has wide application in EEG, but the system performance has a great promotion space due to the limitation of EEG signals. However, the brain-computer interface based on the SEEG signal still lacks a great deal of effective work at present, and research on the brain-computer interface technology in this respect has great improvement.
Disclosure of Invention
The invention aims to overcome the defects of large signal interference, low information transmission rate and the like of the conventional P300 character input brain-computer interface system based on scalp electroencephalogram signals, provides a digital input brain-computer interface system based on SEEG signals, and can realize high-speed and high-accuracy digital input.
In order to achieve the purpose, the technical scheme provided by the invention is as follows: a digital input brain-computer interface system based on segg signals, comprising:
the stimulation presenting module is used for presenting a P300 electroencephalogram mode for stimulating and inducing a human subject;
the signal acquisition module is used for acquiring and amplifying intracranial electroencephalogram signals of a testee;
the model training module is used for training the classifier by using training data to obtain a classifier response model;
the on-line data analysis module is used for analyzing the electroencephalogram data acquired in real time in the on-line process and outputting a result;
wherein, the stimulation presenting module presents the stimulation through a display connected with a computer by a VGA wire; the signal acquisition module is connected with a computer in a bidirectional way through a parallel port line and a network cable, and the model training module and the online data analysis module run on the computer;
respectively carrying out a training data acquisition stage and an on-line test stage on a testee, and training a classifier through the model training module by using data acquired by the data acquisition module after the training data acquisition stage is finished to obtain a classifier response model; in the on-line testing stage, after the on-line data analysis module receives the SEEG electroencephalogram data acquired by the data acquisition module in real time, the classifier response model obtained by the model training module is used for analyzing the SEEG electroencephalogram data in real time and outputting a result.
Furthermore, the display interface of the stimulation presenting module is an n × m matrix interface formed by arranging squares, each square corresponds to a number, and n and m are natural numbers larger than 0;
the working mode of the stimulation presenting module is a single digital flashing mode, namely, after a certain number flashes for a period of time, the next number begins to flash, when the number does not flash, the number is white, and when the number flashes, the number turns to green; the stimulation duration of each flash is preset t1 milliseconds, and the flash interval between two numbers is preset t2 milliseconds; when the stimulation presenting module flickers each time, the stimulation presenting module sends a marking event of a corresponding number to the signal acquisition module through the parallel port, wherein the value of the event is the value of the corresponding number plus n multiplied by m; each flash cycle means that all the n multiplied by m numbers are sequentially flashed once according to a random sequence, and the time of each flash cycle is t3 seconds which are preset; each trial time comprises K rounds of flicker, K is a natural number greater than 0, and the total time is preset t4 seconds;
the signal acquisition module comprises an implanted SEEG electroencephalogram electrode and an electroencephalogram amplification and acquisition device; the SEEG electroencephalogram electrode collects original intracranial electroencephalogram signals through contacts on the electrode; the electroencephalogram amplification and acquisition device is used for amplifying and denoising an original electroencephalogram signal; the signal acquisition module can simultaneously receive the marked event of the corresponding number sent by the stimulation presentation module, the event data and the electroencephalogram data are synchronously acquired, and finally, the processed electroencephalogram data and the event data are recorded in a computer.
Further, the model training module adopts a filter to pre-process SEEG data acquired by each electrode channel of the signal acquisition module, then selects a plurality of sampling points as characteristic points for the electrode signal of each channel, the time length corresponding to the sampling points is t5 milliseconds, so that a characteristic vector corresponding to each electrode channel is acquired, after pre-processing, Bayesian linear discrimination BLDA classification cross validation is performed on the data of each electrode channel, the classification accuracy of each electrode channel on target numbers and non-target numbers is obtained, five electrode channels with the highest classification accuracy are selected, the characteristic vectors corresponding to the selected channels are combined into one characteristic vector, and then Bayesian linear discrimination BLDA classification is performed again by using the integrated characteristic vectors, so that the classifier response of the target numbers and the non-target numbers is obtained;
after receiving a section of data corresponding to each flash number sent back by the signal acquisition module, the online data analysis module performs filtering and sampling point extraction on the selected five channel data to obtain a feature vector of each section of data, and then sends the feature vector to the training classifier to obtain the probability that the number is a target number; after K rounds of flickers are performed, the probability sizes that all n × m numbers are the target numbers are compared, the number with the highest probability is taken as an output, and the result is displayed on a screen.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. the invention uses SEEG signal to construct brain-computer interface for the first time, and the signal has high signal-to-noise ratio and low noise.
2. Compared with the traditional P300 brain-computer interface system constructed based on the EEG signal, the invention uses more than 10 electrodes for data processing, and only uses data of 5 electrode channels for online input experiment, thereby reducing the cost on software and hardware. The calculation method is simplified, the data processing time is shortened, and the possibility of a system with more complex functions is provided for subsequent improvement. Meanwhile, for the implanted brain-computer interface, fewer electrodes mean that the subject is less traumatic.
3. The invention ensures that the online data can be processed in time and the result can be returned quickly, and meanwhile, for the testee using the brain-computer interface for the first time, the character input accuracy rate of more than 90 percent and the maximum information transmission rate of more than 50bits/min can be reached, and compared with the EEG brain-computer interface with the same or similar normal form, the performance is greatly improved.
Drawings
FIG. 1 is a system architecture diagram of the present invention.
FIG. 2 is a schematic diagram of a 4 × 10 matrix interface for inducing P300.
Detailed Description
The present invention will be further described with reference to the following specific examples.
As shown in fig. 1, the present embodiment provides a digital input brain-computer interface system based on segg signals, and the whole system may be divided into a stimulus presentation module, a signal acquisition module, a model training module, and an online data analysis module. The stimulation presentation module is compiled by a C + + program, is presented by a display connected with a computer through a VGA (video graphics array) connection wire and is used for presenting a P300 electroencephalogram mode for stimulating and inducing a testee; the signal acquisition module comprises an implanted SEEG electroencephalogram electrode and an electroencephalogram amplification and acquisition device, is bidirectionally connected with the computer through a parallel port line and a network cable and is used for acquiring and amplifying intracranial electroencephalogram signals of a testee; after the signal acquisition module acquires training data, the model training module carries out classifier training on the data to obtain a classifier response model; and in the process of carrying out an online experiment, the online data analysis module analyzes the data in real time by utilizing a classifier response model obtained by the model training module after receiving the data collected by the signal collection module, and outputs a result.
For the way that the stimulus presentation module induces the P300 waveform, the present embodiment employs a matrix interface of a 4 × 10 matrix in which 40 numbers (01 to 40) are arranged, as shown in fig. 2. The mode of operation of the stimulus presentation module is a single digital flash mode, i.e. a digital flash starts after a certain time when the next digital flash starts. When there is no flicker, the number is white; the numbers at the time of flashing turn green. The stimulation duration of each flash is 100 ms and the flash interval (ISI) between two numbers is 30 ms. Each round of (round) flashing means that all 40 numbers are flashed in sequence according to a random sequence, and the flashing time of one round is 1.2 seconds in total. Each trial (trial) contained 10 rounds of flicker for a total duration of about 12 seconds.
The signal acquisition module comprises an implanted SEEG electroencephalogram electrode and an electroencephalogram amplification and acquisition device; the SEEG electroencephalogram electrode collects original intracranial electroencephalogram signals through contacts on the electrode; the electroencephalogram amplification and acquisition device is used for amplifying and denoising the original electroencephalogram signals, and finally recording the processed electroencephalogram data in a computer.
The model training module adopts a filter to pre-process SEEG data acquired by each electrode channel of the signal acquisition module, then selects a plurality of sampling points as characteristic points for the electrode signal of each channel, the time length corresponding to the sampling points is 600 milliseconds, so that a characteristic vector corresponding to each electrode channel is acquired, after pre-processing, Bayesian linear discrimination BLDA classification cross validation is performed on the data of each electrode channel, the classification accuracy of each electrode channel on target numbers and non-target numbers is obtained, five electrode channels with the highest classification accuracy are selected, the characteristic vectors corresponding to the selected channels are combined into one characteristic vector, and then Bayesian linear discrimination BLDA classification is performed again by using the integrated characteristic vectors, so that the classifier response of the target numbers and the non-target numbers is obtained.
After receiving a section of data corresponding to each flash number transmitted back by the signal acquisition module, the online data analysis module performs filtering and sampling point extraction on the selected five channel data to obtain a feature vector of each section of data, and then transmits the feature vector to the training classifier to obtain the probability that the number is a target number; after 10 rounds of flickers are performed, the probability of all 40 numbers being the target number is compared, the number with the highest probability is used as the output, and the result is displayed on the screen.
Based on the digital input brain-computer interface system, the specific experimental process is as follows:
1) the intracranial electroencephalogram data are collected and amplified through the implanted SEEG electroencephalogram electrode and an electroencephalogram amplification and collection device, waveform detection software matched with electroencephalogram collection equipment is opened, and a direct-current input channel is opened, so that a stimulation program can send markers to an amplifier, time points can be found during offline analysis, and data can be synchronously grabbed during online testing.
2) The electroencephalogram amplification and acquisition device, the stimulation presentation module and the online data analysis module establish a communication relation, and the specific connection mode is as follows:
2.1) opening a stimulus presentation program scutbci.exe in a computer, and automatically operating two subprograms after clicking an operation button on a menu bar, wherein the two subprograms are netreaderntK.exe used for receiving data and netstimNTK.exe used for sending a marker data marker;
2.2) setting a TCP port number on an interface of netreaderntK.exe, establishing connection between the electroencephalogram amplification and acquisition device and the online data analysis module based on a TCP protocol after clicking a connection button, and then receiving SEEG electroencephalogram data sent by the electroencephalogram amplification and acquisition device in real time by the online data analysis module and displaying the received electroencephalogram waveform in a window;
2.3) setting a UDP port number in an interface of netstimNTK.exe, establishing connection between the stimulation presentation module and the electroencephalogram amplification and acquisition device based on a UDP protocol after clicking a sending button, and sending a mark data marker from the stimulation presentation module to the signal acquisition module.
3) Carrying out acquisition of training data:
3.1) clicking a setting button on a menu bar of the scutbci.exe, setting a training trial number (generally 40) in a selection mode of ' training mode ' (train '), clicking a start button on the menu bar after clicking for storage, and starting operation of the stimulation presentation module;
3.2) at the beginning of each trial, randomly prompting a target number, and sending the marked event (the event is the value of the corresponding target number plus 40, and the range is between 41 and 80) of the corresponding number to the signal acquisition module by the stimulation presentation module. The subject needs to look at the flicker of this target number and defaults to the number of flickers. Then the number matrix starts to flash for 10 rounds, and when each number flashes, the stimulation presentation module sends the marked event (the value of the event is between 1 and 40) of the corresponding number to the signal acquisition module. After finishing 40 test times, stopping flashing by the stimulation presenting module, and then closing the record of the signal acquisition module;
3.3) the model training module adopts a filter to preprocess the acquired SEEG electroencephalogram data, and then intercepts 600 milliseconds of data after each flicker, which corresponds to a period (epoch). For each electrode signal channel, after 10 epochs corresponding to the same number are averaged in each trial, a plurality of sampling points are selected as characteristic points for the averaged electrode signals, and thus, a characteristic vector corresponding to each trial is obtained (39 of the characteristic vectors correspond to non-targets which are not stimulated, and 1 of the characteristic vectors corresponds to a target which is stimulated by P300). Then, for the obtained feature vector of each electrode channel, carrying out Bayesian linear discrimination BLDA classification ten-fold cross validation to obtain the classification accuracy of each electrode channel on the target number stimulated by P300 and the non-target number not stimulated, selecting five electrode channels with the highest classification accuracy, and training a model for online input;
3.4) combining the corresponding eigenvectors of the selected five electrode channels into one eigenvector, and then performing Bayesian linear discrimination BLDA classification again by using the integrated eigenvectors of all 40trials (40trials × 39 is 1560 corresponding non-targets, 40trials × 1 is 40 corresponding targets), so as to obtain the classifier response of the target number and the non-target number.
4) Performing an online digital input test:
4.1) clicking a setting button on a menu bar of the scutbci.exe, setting the selection mode to be an online mode (online), setting the trial number of online tests to be generally 40, clicking a starting button on the menu bar after clicking and storing, and starting the operation of the stimulation presentation module;
4.2) we randomly generate a group of numbers in advance, each number is generated from '01' to '40', a total of 40 numbers correspond to 40trials, and the testee is required to input the 40 numbers in turn. For each trial, the subject needs to look at the numbers that need to be entered. Then the number matrix starts 10 rounds of random flickers, each number flickers, and at the same time, the stimulation presentation module sends the marked event (the value of the event is between 1 and 40) of the corresponding number to the signal acquisition module. Meanwhile, the on-line data analysis module automatically captures 600 milliseconds of SEEG electroencephalogram data from each marked event, namely an epoch, and sends the SEEG electroencephalogram data together with the value of the event to the on-line data analysis module. After one round of flashing is finished, 40 numbers are flashed once, and at the same time, the computer collects corresponding 40 epochs. After ten rounds of flashes, the computer collected a total of 400 epochs, 10 for each number.
4.3) after the online data analysis module receives all 400 epochs, carrying out mean processing on 10 epochs corresponding to each number, then selecting feature points of 5 electrode channel data in the same way as offline training data to obtain corresponding feature vectors, then putting 40 feature vectors corresponding to 40 numbers into a classifier to respond, and respectively obtaining the discrimination probability that the corresponding number is the target number. The discrimination probabilities of the 40 digits are compared and the highest corresponding digit is output for display on a screen.
In conclusion, the digital input brain-computer interface system based on the SEEG signal is designed, and the digital input function can be realized by acquiring the electroencephalogram electrode signal with higher intracranial signal quality, decoding and modeling. Compared to a scalp EEG based character input system of the same stimulation paradigm, the system can achieve the same or higher input accuracy and information transfer rate by using a smaller number of electrode channels. This reduces the cost in software and hardware, simplifies the calculation method, shortens the data processing time, and makes it possible to subsequently improve a more sophisticated system. Meanwhile, for the implanted brain-computer interface, fewer electrodes mean that the subject is less traumatic. As most of the testees implanted with the electrodes are epileptic patients, the invention also brings a brand-new interactive experience for the patients, provides an effective way for clinical scientific research, has potential social value and is worthy of popularization.
The above-mentioned embodiments are merely preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, so that the changes in the shape and principle of the present invention should be covered within the protection scope of the present invention.
Claims (3)
1. A digital input brain-computer interface system based on segg signals, comprising:
the stimulation presenting module is used for presenting a P300 electroencephalogram mode for stimulating and inducing a human subject;
the signal acquisition module is used for acquiring and amplifying intracranial electroencephalogram signals of a testee;
the model training module is used for training the classifier by using training data to obtain a classifier response model;
the on-line data analysis module is used for analyzing the electroencephalogram data acquired in real time in the on-line process and outputting a result;
wherein, the stimulation presenting module presents the stimulation through a display connected with a computer by a VGA wire; the signal acquisition module is connected with a computer in a bidirectional way through a parallel port line and a network cable, and the model training module and the online data analysis module run on the computer;
respectively carrying out a training data acquisition stage and an on-line test stage on a testee, and training a classifier through the model training module by using data acquired by the data acquisition module after the training data acquisition stage is finished to obtain a classifier response model; in the on-line testing stage, after the on-line data analysis module receives the SEEG electroencephalogram data acquired by the data acquisition module in real time, the classifier response model obtained by the model training module is used for analyzing the SEEG electroencephalogram data in real time and outputting a result.
2. The SEEG signal-based digital input brain-computer interface system of claim 1, wherein: the display interface of the stimulation presenting module is an n x m matrix interface formed by arranging squares, each square corresponds to a number, and n and m are natural numbers larger than 0;
the working mode of the stimulation presenting module is a single digital flashing mode, namely, after a certain number flashes for a period of time, the next number begins to flash, when the number does not flash, the number is white, and when the number flashes, the number turns to green; the stimulation duration of each flash is preset t1 milliseconds, and the flash interval between two numbers is preset t2 milliseconds; when the stimulation presenting module flickers each time, the stimulation presenting module sends a marking event of a corresponding number to the signal acquisition module through the parallel port, wherein the value of the event is the value of the corresponding number plus n multiplied by m; each flash cycle means that all the n multiplied by m numbers are sequentially flashed once according to a random sequence, and the time of each flash cycle is t3 seconds which are preset; each trial time comprises K rounds of flicker, K is a natural number greater than 0, and the total time is preset t4 seconds;
the signal acquisition module comprises an implanted SEEG electroencephalogram electrode and an electroencephalogram amplification and acquisition device; the SEEG electroencephalogram electrode collects original intracranial electroencephalogram signals through contacts on the electrode; the electroencephalogram amplification and acquisition device is used for amplifying and denoising an original electroencephalogram signal; the signal acquisition module can simultaneously receive the marked event of the corresponding number sent by the stimulation presentation module, the event data and the electroencephalogram data are synchronously acquired, and finally, the processed electroencephalogram data and the event data are recorded in a computer.
3. The SEEG signal-based digital input brain-computer interface system of claim 1, wherein: the model training module adopts a filter to pre-process SEEG data acquired by each electrode channel of the signal acquisition module, then selects a plurality of sampling points as characteristic points for the electrode signal of each channel, the time length corresponding to the sampling points is t5 milliseconds, so that a characteristic vector corresponding to each electrode channel is acquired, after pre-processing, Bayesian linear discrimination BLDA classification cross validation is performed on the data of each electrode channel, the classification accuracy of each electrode channel on target numbers and non-target numbers is obtained, five electrode channels with the highest classification accuracy are selected, the characteristic vectors corresponding to the selected channels are combined into one characteristic vector, and then Bayesian linear discrimination BLDA classification is performed again by using the integrated characteristic vectors, so that the classifier response of the target numbers and the non-target numbers is obtained;
after receiving a section of data corresponding to each flash number sent back by the signal acquisition module, the online data analysis module performs filtering and sampling point extraction on the selected five channel data to obtain a feature vector of each section of data, and then sends the feature vector to the training classifier to obtain the probability that the number is a target number; after K rounds of flickers are performed, the probability sizes that all n × m numbers are the target numbers are compared, the number with the highest probability is taken as an output, and the result is displayed on a screen.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011012679.8A CN112162634A (en) | 2020-09-24 | 2020-09-24 | Digital input brain-computer interface system based on SEEG signal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011012679.8A CN112162634A (en) | 2020-09-24 | 2020-09-24 | Digital input brain-computer interface system based on SEEG signal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112162634A true CN112162634A (en) | 2021-01-01 |
Family
ID=73863606
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011012679.8A Pending CN112162634A (en) | 2020-09-24 | 2020-09-24 | Digital input brain-computer interface system based on SEEG signal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112162634A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113397559A (en) * | 2021-06-17 | 2021-09-17 | 深圳大学 | Stereotactic electroencephalogram analysis method, stereotactic electroencephalogram analysis device, computer equipment and storage medium |
| CN113867533A (en) * | 2021-09-30 | 2021-12-31 | 中国人民解放军战略支援部队信息工程大学 | Multi-brain cooperative brain-computer interface system and video target detection method based on same |
| CN116011506A (en) * | 2023-03-28 | 2023-04-25 | 同心智医科技(北京)有限公司 | Method for constructing stereo electroencephalogram electrode signal decoding model and application thereof |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103488297A (en) * | 2013-09-30 | 2014-01-01 | 华南理工大学 | Online semi-supervising character input system and method based on brain-computer interface |
| CN104127179A (en) * | 2014-04-13 | 2014-11-05 | 北京工业大学 | Electroencephalogram (EEG) feature extraction method based on dominant electrode combination and empirical mode decomposition (EMD) |
| CN108236464A (en) * | 2017-12-29 | 2018-07-03 | 重庆邮电大学 | Feature extracting method and its Detection and Extraction system based on EEG signals |
| CN109924970A (en) * | 2019-04-19 | 2019-06-25 | 李文玲 | The method and terminal device of brain network telecommunication number analysis |
| CN110262658A (en) * | 2019-06-14 | 2019-09-20 | 长春理工大学 | A kind of brain-computer interface character input system and implementation method based on reinforcing attention |
| EP3561645A1 (en) * | 2018-04-25 | 2019-10-30 | Fujitsu Limited | Deep neural network training for application program generation |
| US20200125877A1 (en) * | 2018-10-22 | 2020-04-23 | Future Health Works Ltd. | Computer based object detection within a video or image |
-
2020
- 2020-09-24 CN CN202011012679.8A patent/CN112162634A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103488297A (en) * | 2013-09-30 | 2014-01-01 | 华南理工大学 | Online semi-supervising character input system and method based on brain-computer interface |
| CN104127179A (en) * | 2014-04-13 | 2014-11-05 | 北京工业大学 | Electroencephalogram (EEG) feature extraction method based on dominant electrode combination and empirical mode decomposition (EMD) |
| CN108236464A (en) * | 2017-12-29 | 2018-07-03 | 重庆邮电大学 | Feature extracting method and its Detection and Extraction system based on EEG signals |
| EP3561645A1 (en) * | 2018-04-25 | 2019-10-30 | Fujitsu Limited | Deep neural network training for application program generation |
| US20200125877A1 (en) * | 2018-10-22 | 2020-04-23 | Future Health Works Ltd. | Computer based object detection within a video or image |
| CN109924970A (en) * | 2019-04-19 | 2019-06-25 | 李文玲 | The method and terminal device of brain network telecommunication number analysis |
| CN110262658A (en) * | 2019-06-14 | 2019-09-20 | 长春理工大学 | A kind of brain-computer interface character input system and implementation method based on reinforcing attention |
Non-Patent Citations (2)
| Title |
|---|
| WEICHEN HUANG等: "《A P300-based Brain Computer Interface Using Stereo-electroencephalography Signals》", 《2019 41ST ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY (EMBC)》 * |
| 陈柱兵: "《基于P300的高速字符输入脑机接口研究》", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113397559A (en) * | 2021-06-17 | 2021-09-17 | 深圳大学 | Stereotactic electroencephalogram analysis method, stereotactic electroencephalogram analysis device, computer equipment and storage medium |
| CN113397559B (en) * | 2021-06-17 | 2023-08-18 | 深圳大学 | Stereotactic electroencephalogram analysis method, stereotactic electroencephalogram analysis device, computer equipment and storage medium |
| CN113867533A (en) * | 2021-09-30 | 2021-12-31 | 中国人民解放军战略支援部队信息工程大学 | Multi-brain cooperative brain-computer interface system and video target detection method based on same |
| CN113867533B (en) * | 2021-09-30 | 2023-05-05 | 中国人民解放军战略支援部队信息工程大学 | Multi-brain cooperative brain-computer interface system and video target detection method realized based on same |
| CN116011506A (en) * | 2023-03-28 | 2023-04-25 | 同心智医科技(北京)有限公司 | Method for constructing stereo electroencephalogram electrode signal decoding model and application thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110765920B (en) | Motor imagery classification method based on convolutional neural network | |
| CN112162634A (en) | Digital input brain-computer interface system based on SEEG signal | |
| US20140303511A1 (en) | Single trial detection in encephalography | |
| CN106407733A (en) | Depression risk screening system and method based on virtual reality scene electroencephalogram signal | |
| CN112163518B (en) | Emotion modeling method for emotion monitoring and adjusting system | |
| Pun et al. | Brain-computer interaction research at the Computer Vision and Multimedia Laboratory, University of Geneva | |
| CN108958486A (en) | Mood brain machine interface system based on recognition of face brain electricity N170 ingredient | |
| Szafir et al. | An exploration of the utilization of electroencephalography and neural nets to control robots | |
| Abibullaev et al. | A brute-force CNN model selection for accurate classification of sensorimotor rhythms in BCIs | |
| CN110262658B (en) | Brain-computer interface character input system based on enhanced attention and implementation method | |
| Li et al. | An adaptive P300 model for controlling a humanoid robot with mind | |
| Zhang et al. | Spatial-temporal neural network for P300 detection | |
| Yang et al. | Time-frequency selection in two bipolar channels for improving the classification of motor imagery EEG | |
| WO2024083059A1 (en) | Working memory task magnetoencephalography classification system based on machine learning | |
| CN115392287A (en) | Electroencephalogram signal online self-adaptive classification method based on self-supervision learning | |
| CN114063773A (en) | Cross-individual brain-computer interface system based on convolutional neural network and electroencephalogram big data | |
| More et al. | Using motor imagery and deeping learning for brain-computer interface in video games | |
| Ahmed et al. | A non Invasive Brain-Computer-Interface for Service Robotics | |
| Zhang et al. | Motor imagery EEG classification with self-attention-based convolutional neural network | |
| Bandara et al. | Differentiation of signals generated by eye blinks and mouth clenching in a portable brain computer interface system | |
| Bi et al. | A SSVEP brain-computer interface with the hybrid stimuli of SSVEP and P300 | |
| Mousavi et al. | Motor imagery performance from calibration to online control in EEG-based brain-computer interfaces | |
| Ananthi et al. | A review on-EEG signals by motor imagery based brain computer interface | |
| Li et al. | Exploiting EEG channel correlations in P300 speller paradigm for brain-computer interface | |
| Plechawska-Wójcik et al. | Comparative analysis of two-group supervised classification algorithms in the study of P300-based brain-computer interface |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210101 |
|
| RJ01 | Rejection of invention patent application after publication |