CN109063639A - A kind of method of real-time prediction Brain behavior - Google Patents
A kind of method of real-time prediction Brain behavior Download PDFInfo
- Publication number
- CN109063639A CN109063639A CN201810853377.XA CN201810853377A CN109063639A CN 109063639 A CN109063639 A CN 109063639A CN 201810853377 A CN201810853377 A CN 201810853377A CN 109063639 A CN109063639 A CN 109063639A
- Authority
- CN
- China
- Prior art keywords
- brain
- behavior
- prediction
- space characteristics
- input vector
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses a kind of methods of Brain behavior of prediction in real time, comprising the following steps: pre-processes to brain signal；Using the brain signal that per moment acquires as input vector, space characteristics extraction is carried out to input vector using space characteristics extraction unit, obtains the space characteristics of input vector；Memory, output timing feature are updated to corresponding space characteristics of per moment using timing memory cycling element；Classified using behavior judging unit to temporal aspect, exports the probability of Brain behavior to realize the prediction to Brain behavior.This method extracts the tendency information contained in clock signal by timing memory circular treatment, and the accuracy of prediction Brain behavior is improved with this.
Description
Technical field
The invention belongs to brain brain-computer interface fields, and in particular to a method of prediction Brain behavior in real time.
Background technique
The all life activity of cerebral dominance people, occurs 100,000 kinds of chemical reactions every second, is obtained by sense organ extraneous
Information and by central nervous system pass to access control human body each organ make corresponding reaction.For a long time,
More and more scientists are keen to the research of nervous system, and brain is understood using a variety of research methods to the volume of external event
Ink recorder system.It is desirable to directly be communicated using the movable signal of human thinking with the external world, or even realize to ambient enviroment
Control.
Neuscience studies have shown that brain generation movement consciousness after and movement execute before or subject
After by environmental stimuli, the electrical activity of nervous system can occur to change accordingly.The working principle of brain machine interface system is exactly
Detect the movable variation of this neuroelectricity, and as acting imminent characteristic signal, by these characteristic signals into
Row Classification and Identification tells the movement intention for causing brain Electrical change, computer language is recycled to be programmed, and the thinking people is living
Turn becomes command signal driving external equipment and realizes human brain pair in the case where no muscle and peripheral nerve directly participate in
The control of external environment, and the communication with the external world.Currently, reading out the ongoing various thinking activities of people also by brain electricity
It is unrealistic, but it is possible for distinguishing to certain functions of brain function.
The laboratory research of traditional cell biology etc. is for solving acquisition, processing and processing of the human brain to complex information
And the mechanism of higher cognitive function, like cannot see the wood for the trees.The application of nuroinformatics tool and database, so that I
May be found out from limited experimental data nerve information obtain, processing and integration rule and rule, propose in various thorns
Under the conditions of swashing, the experimental hypothesis of the mathematical model of intracerebral Information procession and with computer simulation intracerebral nerve information network.
Existing Brain behavior detection method, such as Kalman filtering (Kalman Filter), particle filter
Various filtering algorithms such as (Particle Filter), and emerge in large numbers in recent years convolutional neural networks, Boltzmann machine even depth
The method of study, the overwhelming majority are to identify, predict that merely brain is current or next according to current time collected signal
The behavior and motion state at moment, and all data occurred before not making full use of.Minority based on " N-gram " or
The method of person's Markov model, although the sequence occurred before part is utilized, the sequence length utilized is usually all very
Short, 3~9 time points are usually only utilized in " N-gram " length, and " Markov " is even more the length at only 2~3 time points.
Therefore, existing method is difficult to accurately identify, predict that brain is current or the behavior of following generation at present.If energy
The method for enough proposing the real-time prediction Brain behavior of sequence information before can the making full use of of a kind of novel high-precision, that will be
It is very significant.
Summary of the invention
The dependence for not accounting for clock signal for existing prediction technique causes prediction result accuracy is low to ask
Topic, the present invention provides a kind of methods of Brain behavior of prediction in real time to be extracted in clock signal by timing memory circular treatment
The tendency information contained improves the accuracy of prediction Brain behavior with this.
For achieving the above object, the present invention the following technical schemes are provided:
A kind of method of real-time prediction Brain behavior, comprising the following steps:
Brain signal is pre-processed；
Using the brain signal that per moment acquires as input vector, input vector is carried out using space characteristics extraction unit
Space characteristics extract, and obtain the space characteristics of input vector；
Memory, output timing feature are updated to corresponding space characteristics of per moment using timing memory cycling element；
Temporal aspect is judged using behavior judging unit, exports the probability of Brain behavior to realize to Brain behavior
Detection.
The present invention using timing memory cycling element in timing brain signal dependence capture, according to when
The implicit internal association feature of sequence brain signal, realizes the prediction to Brain behavior.
Above-mentioned brain signal includes brain wave (Electroencephalogram, EEG), magneticencephalogram
(magnetoencephalogram, Meg), local field potentials (local field potential, LFP), functional magnetic resonance
(functional magnetic resonance imaging, FMRI) etc. is imaged.
The brain signal of acquisition can have many noises, and be influenced by heartbeat, breathing and body kinematics, therefore need
Brain signal is pre-processed, to purify brain signal.It is described to brain signal carry out pretreatment include: to brain signal
It is filtered, the dynamic correction of noise reduction, head, the processing of heartbeat breath calibration.
Space characteristics extraction unit is mainly used for carrying out the transient brain signal of input feature extraction, reduction input letter
Number dimension, extract result for visualizing, the main brain area spatial distribution and feature explained when various actions occur for brain
Brain map.It for space characteristics, can both be extracted, can also be extracted using all kinds of Mathematical Methods using neural network.Cause
This, it is preferable that in the space characteristics extraction unit, utilize trained perceptron, full Connection Neural Network, convolutional Neural
Network, depth confidence neural network, Boltzmann machine or limitation Boltzmann machine carry out space characteristics extraction to input vector.It is excellent
Selection of land carries out input vector using principal component analysis or Independent Component Analysis in the space characteristics extraction unit
Space characteristics extract.
The dependence that timing memory cycling element is mainly used for lying in continuous sequence is modeled, learning of structure sequence
The Dynamic Changes rule and probabilistic model implied in column is to realize that the update to instantaneous space feature is remembered.Preferably, in institute
It states in timing memory cycling element, is realized using trained Recognition with Recurrent Neural Network or recurrent neural network corresponding to per moment
The update of space characteristics is remembered.
Recognition with Recurrent Neural Network and recurrent neural network can select to update and remember, to mention while carrying out feature extraction
The Dynamic Changes rule for taking and recording signal in clock signal sequence, provides effective data base for subsequent judgement Brain behavior
Plinth.
Preferably, in the behavior judging unit, using softmax classifier or support vector machines to temporal aspect into
Row classification.
Further, the space characteristics extraction unit uses trained convolutional neural networks, and the timing memory follows
Ring element uses Recognition with Recurrent Neural Network, and the behavior judging unit uses softmax classifier.Experiment proves that passing through the group
It closes, 90% is up to the predictablity rate of Brain behavior, it is higher using non-sequential classifier prediction Brain behavior accuracy rate than existing
20%.
The device have the advantages that are as follows:
Method provided by the invention has fully considered the tendency information contained in timing brain signal, the knowledge to Brain behavior
Not/prediction Average Accuracy is higher than about 20 percentage points of non-sequential classifier, substantially increases the knowledge of Brain behavior up to 90%
Not/precision of prediction.
Detailed description of the invention
Fig. 1 is the realization system structure diagram of real-time prediction Brain behavior method provided by the invention；
Fig. 2 is the working principle diagram of real-time prediction Brain behavior method provided by the invention；
Be 1. input signal in Fig. 1 and Fig. 2, be 2. signal processing unit, be 3. space characteristics extraction unit, 4. for when
Sequence remembers cycling element, is 5. behavior judging unit, is 6. output result.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention more comprehensible, with reference to the accompanying drawings and embodiments to this
Invention is described in further detail.It should be appreciated that the specific embodiments described herein are only used to explain the present invention,
And the scope of protection of the present invention is not limited.
Fig. 1 is the realization system structure diagram of real-time prediction Brain behavior method provided by the invention.As shown in Figure 1,
The system includes that successively cascade signal processing unit, space characteristics extraction unit, timing memory cycling element, behavior determine list
Member.
From each station acquisition of brain to brain signal sequence, temporally dimension is split, i.e., same time point adopts
The brain signal of each position collected constitutes input vector.Here, the input vector of a certain moment t is expressed as VtIf in total
Acquiring the data of T time length altogether in N number of position, then input data matrix dimension is M=N × T, wherein | V |=N.
Original brain signal is filtered by signal processing unit, noise reduction, head move correction, at heartbeat breath calibration
Reason, to purify brain signal.After brain signal pretreatment, T input vector is input to the progress of space characteristics extraction unit one by one
Space characteristics extract, and since each input vector is directed to the brain signal data of single point in time, are consequently belonging to space model
The feature extraction enclosed, not temporal signatures extract.
Spatial signature vectors by extraction are input to timing memory cycling element.The unit is preserved over processed
The recall info of all spatial signature vectors, when there is the input of new spatial signature vectors, timing memory cycling element can be autonomous
Decide whether to update memory (why), how to update memory (how) and updates which memory (what).Therefore, according to current
Input space feature, and the memory that input space feature is remained before, behavior judging unit can be made accurately
Behavior determines.
In this method, Brain behavior feature recognition principle are as follows: according to sequence x has occurred before1,x2,…,xt-1, and work as
Preceding input xt, seek the behavior y of t moment maximum probability generationt, ytThe probability of happening of behavior is expressed as p (yt=1 | x1,x2,…,
xt)。
The prediction process of this method is the process of a searching maximum probability behavior, Brain behavior prediction principle are as follows: according to
Sequence x has occurred before1,x2,…,xt, seek the behavior y of t+1 moment maximum probability generationt+1, yt+1The probability of happening table of behavior
It is shown as p (yt+1=1 | x1,x2,…,xt), and the probability of existing method is expressed as p (yt+1=1 | xt+1)。
Fig. 2 is expanded view of the Fig. 1 in time dimension, from fig. 2 it can be seen that timing memory cycling element is in time dimension
On be associated, the information of moment characteristic vector sequence before can all be saved in current time, and sentence for final result
It is fixed that foundation is provided.
In the present embodiment, space characteristics extraction unit is using the full Connection Neural Network for including 512 neurons, timing
Remember cycling element use double-layer structure Recognition with Recurrent Neural Network, wherein every layer of Recognition with Recurrent Neural Network include 128 LSTM or
GRU circulation memory neuron, behavior judging unit use Softmax classifier.According to currently to cerebral function connection map
Research Literature, the unique functional network connection quantity of brain is about more than ten to tens under various task states, in diseases such as epilepsies
Under diseased state, after some time it is possible to reach 100 or so.For excessively complete redundant representation, this test is chosen in space characteristics extraction unit
512 neurons are to extract 512 feature functionalities connection networks.In order to determine circulation memory in timing memory cycling element
The quantity of neuron, test is attempted in 16~256 ranges, and finally when comprising 128 units, system determines result
Most preferably.This system in the training process, the creative training method using random fragment, i.e., from the random order of time series
The sequence fragment for setting place's interception random-length (usually the 30%~50% of sequence total length), inputs this method for this segment and mentions
In structure out.The method can greatly increase the diversity of sequence samples, between reduction individual caused by similar tasks design
It influences, the reduction over-fitting of high degree.Meanwhile (ratio is kept using dropout strategy to timing memory cycling element
For 0.5), to further decrease over-fitting.
Technical solution of the present invention and beneficial effect is described in detail in above-described specific embodiment, Ying Li
Solution is not intended to restrict the invention the foregoing is merely presently most preferred embodiment of the invention, all in principle model of the invention
Interior done any modification, supplementary, and equivalent replacement etc. are enclosed, should all be included in the protection scope of the present invention.
Claims (8)
1. a kind of method of the Brain behavior of prediction in real time, comprising the following steps:
Brain signal is pre-processed；
Using the brain signal that per moment acquires as input vector, space is carried out to input vector using space characteristics extraction unit
Feature extraction obtains the space characteristics of input vector；
Memory, output timing feature are updated to corresponding space characteristics of per moment using timing memory cycling element；
Classified using behavior judging unit to temporal aspect, exports the probability of Brain behavior to realize to the pre- of Brain behavior
It surveys.
2. the method for the Brain behavior of prediction in real time as described in claim 1, which is characterized in that described to be carried out in advance to brain signal
Processing includes:
Brain signal is filtered, the dynamic correction of noise reduction, head, the processing of heartbeat breath calibration.
3. the method for the Brain behavior of prediction in real time as described in claim 1, which is characterized in that extracted in the space characteristics single
In member, trained perceptron, full Connection Neural Network, convolutional neural networks, depth confidence neural network, Boltzmann are utilized
Machine or limitation Boltzmann machine carry out space characteristics extraction to input vector.
4. the method for the Brain behavior of prediction in real time as described in claim 1, which is characterized in that extracted in the space characteristics single
In member, space characteristics extraction is carried out to input vector using principal component analysis or Independent Component Analysis.
5. the method for the Brain behavior of prediction in real time as described in claim 1, which is characterized in that recycled in the timing memory single
In member, realize that the update to corresponding space characteristics of per moment is remembered using trained Recognition with Recurrent Neural Network or recurrent neural network
Recall.
6. the method for the Brain behavior of prediction in real time as described in claim 1, which is characterized in that in the behavior judging unit
In, classified using softmax classifier or support vector machines to temporal aspect.
7. the method for the Brain behavior of prediction in real time as described in claim 1, which is characterized in that the space characteristics extraction unit
Using trained convolutional neural networks, the timing memory cycling element uses Recognition with Recurrent Neural Network, and the behavior determines single
Member uses softmax classifier.
8. the method for real-time prediction Brain behavior as described in any one of claims 1 to 7, which is characterized in that the brain letter
Number include brain wave, magneticencephalogram, local field potentials, functional magnetic resonance imaging.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810853377.XA CN109063639A (en) | 2018-07-30 | 2018-07-30 | A kind of method of real-time prediction Brain behavior |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810853377.XA CN109063639A (en) | 2018-07-30 | 2018-07-30 | A kind of method of real-time prediction Brain behavior |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109063639A true CN109063639A (en) | 2018-12-21 |
Family
ID=64831916
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810853377.XA Pending CN109063639A (en) | 2018-07-30 | 2018-07-30 | A kind of method of real-time prediction Brain behavior |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109063639A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109820507A (en) * | 2019-03-25 | 2019-05-31 | 钟霁媛 | Disease aided diagnosis method and device |
| CN110292380A (en) * | 2019-07-02 | 2019-10-01 | 重庆大学 | A kind of disease diagnosing system based on GRU Recognition with Recurrent Neural Network |
| CN110393525A (en) * | 2019-06-18 | 2019-11-01 | 浙江大学 | A kind of brain activity detection method based on deep-cycle self-encoding encoder |
| CN111248927A (en) * | 2020-01-17 | 2020-06-09 | 北京师范大学 | Method and system for predicting specific interpersonal relationship through cerebral blood oxygen signals |
| CN111728590A (en) * | 2020-06-30 | 2020-10-02 | 中国人民解放军国防科技大学 | Individual cognitive ability prediction method and system based on dynamic function connection |
| CN111950455A (en) * | 2020-08-12 | 2020-11-17 | 重庆邮电大学 | Motion imagery electroencephalogram characteristic identification method based on LFFCNN-GRU algorithm model |
| CN113499524A (en) * | 2021-07-23 | 2021-10-15 | 华南理工大学 | Auxiliary rehabilitation training system using motor imagery electroencephalogram detection |
| CN113948189A (en) * | 2021-12-22 | 2022-01-18 | 北京航空航天大学杭州创新研究院 | MEG source positioning method based on GRU neural network |
| CN113951896A (en) * | 2021-09-10 | 2022-01-21 | 之江实验室 | Brain-computer interface decoding method based on intracranial brain electricity and scalp brain electricity fusion |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1744073A (en) * | 2005-09-26 | 2006-03-08 | 天津大学 | Method for extracting imagination action poteutial utilizing rpplet nerve net |
| CN105654135A (en) * | 2015-12-30 | 2016-06-08 | 成都数联铭品科技有限公司 | Image character sequence recognition system based on recurrent neural network |
| CN106503616A (en) * | 2016-09-20 | 2017-03-15 | 北京工业大学 | A kind of Mental imagery Method of EEG signals classification of the learning machine that transfinited based on layering |
| CN107625521A (en) * | 2017-09-14 | 2018-01-26 | 华东师范大学 | The multilayer modeling method for being used to assess memory dynamic change based on eeg data |
| CN107961007A (en) * | 2018-01-05 | 2018-04-27 | 重庆邮电大学 | A kind of electroencephalogramrecognition recognition method of combination convolutional neural networks and long memory network in short-term |
-
2018
- 2018-07-30 CN CN201810853377.XA patent/CN109063639A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1744073A (en) * | 2005-09-26 | 2006-03-08 | 天津大学 | Method for extracting imagination action poteutial utilizing rpplet nerve net |
| CN105654135A (en) * | 2015-12-30 | 2016-06-08 | 成都数联铭品科技有限公司 | Image character sequence recognition system based on recurrent neural network |
| CN106503616A (en) * | 2016-09-20 | 2017-03-15 | 北京工业大学 | A kind of Mental imagery Method of EEG signals classification of the learning machine that transfinited based on layering |
| CN107625521A (en) * | 2017-09-14 | 2018-01-26 | 华东师范大学 | The multilayer modeling method for being used to assess memory dynamic change based on eeg data |
| CN107961007A (en) * | 2018-01-05 | 2018-04-27 | 重庆邮电大学 | A kind of electroencephalogramrecognition recognition method of combination convolutional neural networks and long memory network in short-term |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109820507A (en) * | 2019-03-25 | 2019-05-31 | 钟霁媛 | Disease aided diagnosis method and device |
| CN109820507B (en) * | 2019-03-25 | 2024-03-08 | 钟霁媛 | Disease auxiliary diagnosis method and device |
| CN110393525A (en) * | 2019-06-18 | 2019-11-01 | 浙江大学 | A kind of brain activity detection method based on deep-cycle self-encoding encoder |
| CN110393525B (en) * | 2019-06-18 | 2020-12-15 | 浙江大学 | Brain activity detection method based on deep cycle self-encoder |
| CN110292380A (en) * | 2019-07-02 | 2019-10-01 | 重庆大学 | A kind of disease diagnosing system based on GRU Recognition with Recurrent Neural Network |
| CN111248927A (en) * | 2020-01-17 | 2020-06-09 | 北京师范大学 | Method and system for predicting specific interpersonal relationship through cerebral blood oxygen signals |
| CN111728590A (en) * | 2020-06-30 | 2020-10-02 | 中国人民解放军国防科技大学 | Individual cognitive ability prediction method and system based on dynamic function connection |
| CN111950455A (en) * | 2020-08-12 | 2020-11-17 | 重庆邮电大学 | Motion imagery electroencephalogram characteristic identification method based on LFFCNN-GRU algorithm model |
| CN111950455B (en) * | 2020-08-12 | 2022-03-22 | 重庆邮电大学 | Motion imagery electroencephalogram characteristic identification method based on LFFCNN-GRU algorithm model |
| CN113499524A (en) * | 2021-07-23 | 2021-10-15 | 华南理工大学 | Auxiliary rehabilitation training system using motor imagery electroencephalogram detection |
| CN113951896A (en) * | 2021-09-10 | 2022-01-21 | 之江实验室 | Brain-computer interface decoding method based on intracranial brain electricity and scalp brain electricity fusion |
| CN113948189A (en) * | 2021-12-22 | 2022-01-18 | 北京航空航天大学杭州创新研究院 | MEG source positioning method based on GRU neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109063639A (en) | A kind of method of real-time prediction Brain behavior | |
| CN108304917B (en) | P300 signal detection method based on LSTM network | |
| CN113693613B (en) | Electroencephalogram signal classification method, electroencephalogram signal classification device, computer equipment and storage medium | |
| Cecotti et al. | Single-trial classification of event-related potentials in rapid serial visual presentation tasks using supervised spatial filtering | |
| CN112861604B (en) | Myoelectric action recognition and control method irrelevant to user | |
| CN110070105B (en) | Electroencephalogram emotion recognition method and system based on meta-learning example rapid screening | |
| Pinzón-Arenas et al. | Convolutional neural network for hand gesture recognition using 8 different EMG signals | |
| Doborjeh et al. | Classification and segmentation of fMRI spatio-temporal brain data with a NeuCube evolving spiking neural network model | |
| CN111476158A (en) | Multi-channel physiological signal somatosensory gesture recognition method based on PSO-PCA-SVM | |
| CN108874137B (en) | General model for gesture action intention detection based on electroencephalogram signals | |
| CN109948427A (en) | A kind of idea recognition methods based on long memory models in short-term | |
| CN111772629B (en) | Brain cognitive skill transplanting method | |
| Du et al. | A product fuzzy convolutional network for detecting driving fatigue | |
| Bablani et al. | Lie detection using fuzzy ensemble approach with novel defuzzification method for classification of EEG signals | |
| Zhang et al. | An amplitudes-perturbation data augmentation method in convolutional neural networks for EEG decoding | |
| Janapati et al. | Signal processing algorithms based on evolutionary optimization techniques in the BCI: A review | |
| Chuang et al. | Driver's cognitive state classification toward brain computer interface via using a generalized and supervised technology | |
| Lv et al. | Cognitive computing for brain–computer interface-based computational social digital twins systems | |
| CN112244877B (en) | Brain intention identification method and system based on brain-computer interface | |
| Lu et al. | A convolutional neural network based on batch normalization and residual block for P300 signal detection of P300-speller system | |
| CN114916928B (en) | Human body posture multichannel convolutional neural network detection method | |
| Zheng et al. | A Spiking Neural Network based on Neural Manifold for Augmenting Intracortical Brain-Computer Interface Data | |
| Chuang et al. | Automatic design for independent component analysis based brain-computer interfacing | |
| De et al. | EEG-Based Intelligence Quotient Assessment Using 1D Convolutional Neural Network | |
| Bako et al. | Neuromorphic neural network parallelization on CUDA compatible GPU for EEG signal classification |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181221 |