CN109805928A - A kind of cerebral apoplexy Method of EEG signals classification and system - Google Patents
A kind of cerebral apoplexy Method of EEG signals classification and system Download PDFInfo
- Publication number
- CN109805928A CN109805928A CN201910278823.3A CN201910278823A CN109805928A CN 109805928 A CN109805928 A CN 109805928A CN 201910278823 A CN201910278823 A CN 201910278823A CN 109805928 A CN109805928 A CN 109805928A
- Authority
- CN
- China
- Prior art keywords
- cerebral apoplexy
- eeg signals
- sorted
- cerebral
- sample
- 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
Abstract
The invention discloses a kind of cerebral apoplexy Method of EEG signals classification and systems.This method comprises: obtaining cerebral apoplexy EEG signals to be sorted；The cerebral apoplexy EEG signals to be sorted are pre-processed；Layered shaping is carried out to pretreated cerebral apoplexy EEG signals to be sorted；The hierarchical fuzzy entropy characteristic value of cerebral apoplexy EEG signals to be sorted after extracting layering；Obtain disaggregated model；Classified by the disaggregated model to the cerebral apoplexy EEG signals to be sorted according to the hierarchical fuzzy entropy characteristic value.The present invention carries out the feature extraction of hierarchical fuzzy entropy using EEG signals, and is classified with support vector machines, obtains higher discrimination, provides certain reference for clinical analysis diagnosis.
Description
Technical field
The present invention relates to eeg signal classification fields, more particularly to a kind of cerebral apoplexy Method of EEG signals classification and are
System.
Background technique
Cerebral apoplexy is that a kind of brain blood circulation generates disease caused by obstacle, has seriously endangered the health of the mankind.It is sent out
On the one hand Anttdisease Mechanism is since the angiorrhoxis of brain leads to brain disorder caused by bleeding, be on the other hand by the cerebrovascular
It is narrow to lead to brain disorder caused by ischemic.Cerebral apoplexy is generally divided into two types, and one is the blood supplies as caused by thrombus
Insufficient and initiation cerebral arterial thrombosis, also known as cerebral infarction, another kind are the hemorrhagic apoplexies as caused by bleeding, account for whole brains
The 20%~30% of stroke, also known as cerebral hemorrhage.
In recent years, numerous scholars expand multinomial research to the eeg signal classification of cerebral apoplexy.Research direction mainly includes
Two major classes: one is inquire into the feature extracting method for being suitable for EEG signal；Another kind is to inquire into suitable classification method with to mentioning
The feature taken is classified.In terms of feature extraction, autoregression model, Hilbert-Huang transform, Lyapunov index etc. are answered
For in the feature extraction of EEG signals.Common Method of EEG signals classification then specifically includes that artificial neural network, arest neighbors
Classifier, support vector machines etc..In addition, also proposed the analysis method of some entropys in recent years: arrangement entropy, approximate entropy, Sample Entropy,
Fuzzy entropy etc..In dynamic system, what entropy indicated is the generation rate of new information, and the probability for generating new information is bigger, sequence
Complexity is bigger.2011, Jiang etc. proposed the concept of layered entropy again, by layering, while considering the low frequency and height of signal
Frequency ingredient extracts the sample entropy information of different sequence of layer.The method of these entropys provides new think of for analysis cerebral apoplexy EEG signals
Road.At present in the sort research about cerebral apoplexy EEG signals, classification, the brain soldier of cerebral apoplexy patient Mental imagery are focused primarily upon
In the classification of middle patient's thinking mistake area and the sort research of cerebral apoplexy patient emotion, to cerebral apoplexy patient cerebral infarction and cerebral hemorrhage
Sort research it is still seldom.But effectively carry out cerebral infarction and cerebral hemorrhage sort research, to improve cerebral apoplexy accuracy rate of diagnosis,
It reduces cerebral apoplexy disability rate and case fatality rate has great importance.
Summary of the invention
The object of the present invention is to provide a kind of cerebral apoplexy Method of EEG signals classification and systems, are divided using EEG signals
The feature extraction of layer fuzzy entropy, and classified with support vector machines, higher discrimination is obtained, is provided for clinical analysis diagnosis
Certain reference.
To achieve the above object, the present invention provides following schemes:
A kind of cerebral apoplexy Method of EEG signals classification, which comprises
Obtain cerebral apoplexy EEG signals to be sorted；
The cerebral apoplexy EEG signals to be sorted are pre-processed；
Layered shaping is carried out to pretreated cerebral apoplexy EEG signals to be sorted；
The hierarchical fuzzy entropy characteristic value of cerebral apoplexy EEG signals to be sorted after extracting layering；
Obtain disaggregated model；
According to the hierarchical fuzzy entropy characteristic value, by the disaggregated model, to the cerebral apoplexy brain telecommunications to be sorted
Number classify.
Optionally, described that the cerebral apoplexy EEG signals to be sorted are pre-processed, it specifically includes:
Low-pass filtering is carried out to the cerebral apoplexy EEG signals to be sorted；
Segment processing is first carried out to filtered cerebral apoplexy EEG signals.
Optionally, in the acquisition disaggregated model, before further include:
Obtain supporting vector machine model and training sample；The training sample is cerebral apoplexy brain electricity sample signal, described
Cerebral apoplexy brain electricity sample signal includes cerebral infarction sample signal and cerebral hemorrhage sample signal；
The training sample is pre-processed；
Layered shaping is carried out to pretreated training sample；
The fuzzy entropy characteristic value of training sample after extracting layered shaping；
By the sample hierarchical fuzzy entropy characteristic value training supporting vector machine model, disaggregated model is obtained.
Optionally, further includes: dimensionality reduction is carried out to hierarchical fuzzy entropy characteristic value by Principal Component Analysis.
The present invention also provides a kind of cerebral apoplexy eeg signal classification system, the system comprises:
First obtains module, for obtaining cerebral apoplexy EEG signals to be sorted；
First preprocessing module, for being pre-processed to the cerebral apoplexy EEG signals to be sorted；
First layer module, for carrying out layered shaping to pretreated cerebral apoplexy EEG signals to be sorted
First extraction module, for extracting the hierarchical fuzzy entropy feature of the cerebral apoplexy EEG signals to be sorted after being layered
Value；
Model obtains module, for obtaining disaggregated model；
Categorization module is used for according to the hierarchical fuzzy entropy characteristic value, by the disaggregated model, to described to be sorted
Cerebral apoplexy EEG signals are classified.
Optionally, first preprocessing module, specifically includes:
Filter unit, for carrying out low-pass filtering to the cerebral apoplexy EEG signals to be sorted；
Segmenting unit, for first carrying out segment processing to filtered cerebral apoplexy EEG signals.
Optionally, the categorizing system further include:
Second obtains module, for obtaining supporting vector machine model and training sample；The training sample is cerebral apoplexy
Brain electricity sample signal, the cerebral apoplexy brain electricity sample signal include cerebral infarction sample signal and cerebral hemorrhage sample signal；
Second preprocessing module, for being pre-processed to the training sample；
Second hierarchical block, for carrying out layered shaping to pretreated training sample
Second extraction module, for extracting the fuzzy entropy characteristic value of the training sample after layered shaping；
Training module, for being classified by the sample hierarchical fuzzy entropy characteristic value training supporting vector machine model
Model.
Optionally, the categorizing system further include:
Dimensionality reduction module, for carrying out dimensionality reduction to hierarchical fuzzy entropy characteristic value by Principal Component Analysis.
Compared with prior art, the present invention has following technical effect that the present invention to the cerebral apoplexy brain electricity to be sorted
Signal is pre-processed；Extract the hierarchical fuzzy entropy characteristic value of pretreated cerebral apoplexy EEG signals to be sorted；According to institute
Hierarchical fuzzy entropy characteristic value is stated, by the disaggregated model, is classified to the cerebral apoplexy EEG signals to be sorted.It utilizes
EEG signals carry out the feature extraction of hierarchical fuzzy entropy, and are classified with support vector machines, have obtained higher discrimination, have been
Clinical analysis diagnosis provides certain reference.
Detailed description of the invention
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the flow chart of cerebral apoplexy of embodiment of the present invention Method of EEG signals classification；
Fig. 2 is the distribution schematic diagram that cerebral hemorrhage of the embodiment of the present invention and cerebral infarction EEG signals arrange entropy characteristic value；
Fig. 3 is the distribution schematic diagram of cerebral hemorrhage of the embodiment of the present invention and cerebral infarction EEG signals approximate entropy characteristic value；
Fig. 4 is the distribution schematic diagram of cerebral hemorrhage of the embodiment of the present invention and cerebral infarction EEG signals Sample Entropy characteristic value；
Fig. 5 is the distribution schematic diagram of cerebral hemorrhage of the embodiment of the present invention and cerebral infarction EEG signals fuzzy entropy characteristic value；
Fig. 6 is the structural block diagram of cerebral apoplexy of embodiment of the present invention eeg signal classification system.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The object of the present invention is to provide a kind of cerebral apoplexy Method of EEG signals classification and systems, are divided using EEG signals
The feature extraction of layer fuzzy entropy, and classified with support vector machines, higher discrimination is obtained, is provided for clinical analysis diagnosis
Certain reference.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing and specific real
Applying mode, the present invention is described in further detail.
The principle of cerebral apoplexy Method of EEG signals classification provided by the invention are as follows: original EEG signals pass through low-pass filtering
Device retains 35Hz band limits below, then which is divided into the sample set of every section of 1000 data points；To sample
Collection is decomposed to obtain high fdrequency component and low frequency component with wavelet package transforms, and the fuzzy entropy for extracting high frequency and low frequency component respectively is special
Value indicative use Principal Component Analysis to carry out dimensionality reduction after as feature vector, and using algorithm of support vector machine progress disaggregated model instruction
Practice, and classify to test set, obtains final classification result.
As shown in Figure 1, cerebral apoplexy Method of EEG signals classification the following steps are included:
Step 101: obtaining cerebral apoplexy EEG signals to be sorted.
Step 102: the cerebral apoplexy EEG signals to be sorted are pre-processed.To the cerebral apoplexy brain to be sorted
Electric signal carries out low-pass filtering；Segment processing is first carried out to filtered cerebral apoplexy EEG signals.
Step 103: layered shaping is carried out to pretreated cerebral apoplexy EEG signals to be sorted
Step 104: the hierarchical fuzzy entropy characteristic value of the cerebral apoplexy EEG signals to be sorted after extracting layering.
Step 105: obtaining disaggregated model.
Step 106: according to the hierarchical fuzzy entropy characteristic value, by the disaggregated model, to the brain soldier to be sorted
Middle EEG signals are classified.
Before step 105, further includes:
Obtain supporting vector machine model and training sample；The training sample is cerebral apoplexy brain electricity sample signal, described
Cerebral apoplexy brain electricity sample signal includes cerebral infarction sample signal and cerebral hemorrhage sample signal；
The training sample is pre-processed；
Extract the sample hierarchical fuzzy entropy characteristic value of pretreated training sample；
By the sample hierarchical fuzzy entropy characteristic value training supporting vector machine model, disaggregated model is obtained.
For the comparison for carrying out experimental result, it is also extracted approximate entropy, Sample Entropy, arrangement entropy respectively and is divided as characteristic value
Class, and obtained the result of four kinds of entropy characteristic value layering front and backs.
The distribution situation for being layered preceding four kinds of entropy characteristic values is intuitively illustrated present invention employs box figure, such as Fig. 2-Fig. 5 institute
Show.
Arrangement entropy, approximate entropy, Sample Entropy and the mould of cerebral hemorrhage and cerebral infarction EEG signals are calculated separately with 346 sample sets
The value for pasting entropy, finds out their maximum value, minimum value etc., obtained experimental data is indicated with box figure.
Box figure is mainly made of five numerical points, from top to bottom successively are as follows: minimum value, lower quartile, and median, upper four
Quantile, maximum value.Wherein lower quartile and upper quartile form an empty packet, and median is divided into two box.
Upper quartile is counted between minimum value and is respectively connected with an extension line with lower quartile between maximum value."+" generation in figure
Table outlier, the purpose of the independent remittance abroad of outlier are the stability of guarantee global feature, therefore data will not deviate, and in box figure
The two-stage of extension line is modified as minimum (big) observation, minimum (big) observation be empirically set as down (on) quartile subtracts
(adding) 1.5 times of interquartile ranges from.Box figure can not only help us intuitively to identify in sample data when analyzing data
Exceptional value (outlier), and by the length of observation box, the shape of upper and lower chamber and the length of extension line can be effective
The dispersion degree and deviation of judgement sample data.
Fig. 2 to Fig. 5 is respectively the arrangement entropy of cerebral hemorrhage and two class EEG signals of cerebral infarction, approximate entropy, Sample Entropy and fuzzy
The box figure of entropy.What the ordinate of each figure indicated is amplitude, and what abscissa indicated is lead number, and wherein odd number is cerebral hemorrhage phase
The box of lead is answered, even number is the box of cerebral infarction Correspondence lead.It can be seen that for arranging entropy from following figure,
Cerebral hemorrhage and the arrangement entropy feature Distribution value of 8 leads of Patients with Cerebral Infarction EEG signals all compared with concentrate, relative to other three kinds
Exceptional value is relatively more for situation；For approximate entropy and Sample Entropy, cerebral hemorrhage and 8 leads of Patients with Cerebral Infarction EEG signals
Entropy feature Distribution value it is more similar, be distributed it is more dispersed；For fuzzy entropy, cerebral hemorrhage and Patients with Cerebral Infarction EEG signals
The fuzzy entropy feature Distribution value of 8 leads disperses the most, and exceptional value is fewer for other three kinds of situations.
Evaluation index
Evaluation index of the invention uses the common evaluation index of two classification problems: sensibility Se (Sensitivity), special
Anisotropic Sp (Specificity) and accuracy rate Acc (Accuracy), are defined as
In above formula TP, FN, TN, FP respectively represent be classified model prediction be positive class positive class sample number, be classified model
Prediction be negative class positive class sample number, be classified model prediction be negative class negative class sample number and be classified model prediction and be
The negative class sample number of positive class, the value of Se, Sp, Acc show that more greatly classifying quality is better.Positive class sample herein is cerebral hemorrhage sample
This collection, negative class sample are cerebral infarction sample sets.
Experimental result
(1) four kinds of the arrangement entropy, approximate entropy, Sample Entropy and fuzzy entropy entropy characteristic values of patients with cerebral apoplexy EEG signals are extracted simultaneously
Carry out that sorted experimental result is shown in the following table 1.
The experimental result of the cerebral apoplexy classification of 1 four kinds of entropy characteristic values of table
(2) hierarchal arrangement entropy, layered approximate entropy, stratified sample entropy and the hierarchical fuzzy of patients with cerebral apoplexy EEG signals are extracted
Four kinds of entropy characteristic values of entropy simultaneously carry out that sorted experimental result is shown in the following table 2.
The experimental result of the cerebral apoplexy classification of 2 four kinds of layered entropy characteristic values of table
| Experimental program two | Feature vector | Sensibility Se/% | Specific Sp/% | Accuracy rate Acc/% |
| 1 | Hierarchal arrangement entropy | 50.88% | 66.15% | 59.02% |
| 2 | Layered approximate entropy | 100.00% | 92.31% | 95.90% |
| 3 | Stratified sample entropy | 85.96% | 93.85% | 90.16% |
| 4 | Hierarchical fuzzy entropy | 96.49% | 96.92% | 96.72% |
To sum up (1) and (2) it is found that four kinds of entropy characteristic values classified obtained by accuracy rate Acc all 65% or so, and divide
Entropy characteristic value before layer is compared, and the approximate entropy, Sample Entropy and fuzzy entropy after layering can efficiently differentiate cerebral infarction and cerebral hemorrhage
The EEG signals of patient, their physical significance is close, the probability size of expression time series generation new information, generates new letter
The probability of breath is bigger, and the complexity of sequence is bigger.The SVM accuracy rate Acc of hierarchical fuzzy entropy is best, followed by layered approximate
Entropy, stratified sample entropy.Fuzzy membership function is used to replace in Sample Entropy Heavside function as similarity measurement in fuzzy entropy,
The accuracy rate Acc of cerebral apoplexy EEG signals can be effectively improved.
The specific embodiment provided according to the present invention, the invention discloses following technical effects: the present invention is to described wait divide
The cerebral apoplexy EEG signals of class are pre-processed；Extract the hierarchical fuzzy entropy of pretreated cerebral apoplexy EEG signals to be sorted
Characteristic value；According to the hierarchical fuzzy entropy characteristic value, by the disaggregated model, to the cerebral apoplexy EEG signals to be sorted
Classify.The feature extraction that hierarchical fuzzy entropy is carried out using EEG signals, and is classified with support vector machines, obtained compared with
High discrimination provides certain reference for clinical analysis diagnosis.
As shown in fig. 6, the present invention also provides a kind of cerebral apoplexy eeg signal classification system, the system comprises:
First obtains module 601, for obtaining cerebral apoplexy EEG signals to be sorted.
First preprocessing module 602, for being pre-processed to the cerebral apoplexy EEG signals to be sorted.
First preprocessing module 602, specifically includes:
Filter unit, for carrying out low-pass filtering to the cerebral apoplexy EEG signals to be sorted；
Segmenting unit, for first carrying out segment processing to filtered cerebral apoplexy EEG signals.
First layer module 603, for carrying out layered shaping to pretreated cerebral apoplexy EEG signals to be sorted
First extraction module 604, the hierarchical fuzzy entropy for extracting the cerebral apoplexy EEG signals to be sorted after being layered are special
Value indicative.
Model obtains module 605, for obtaining disaggregated model.
Categorization module 606 is used for according to the hierarchical fuzzy entropy characteristic value, by the disaggregated model, to described wait divide
The cerebral apoplexy EEG signals of class are classified.
The categorizing system further include:
Second obtains module, for obtaining supporting vector machine model and training sample；The training sample is cerebral apoplexy
Brain electricity sample signal, the cerebral apoplexy brain electricity sample signal include cerebral infarction sample signal and cerebral hemorrhage sample signal；
Second preprocessing module, for being pre-processed to the training sample；
Second extraction module, for extracting the sample hierarchical fuzzy entropy characteristic value of pretreated training sample；
Training module, for being classified by the sample hierarchical fuzzy entropy characteristic value training supporting vector machine model
Model.
The categorizing system further include:
Dimensionality reduction module, for carrying out dimensionality reduction to hierarchical fuzzy entropy characteristic value by Principal Component Analysis.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with other
The difference of embodiment, the same or similar parts in each embodiment may refer to each other.For system disclosed in embodiment
For, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is said referring to method part
It is bright.
Used herein a specific example illustrates the principle and implementation of the invention, and above embodiments are said
It is bright to be merely used to help understand method and its core concept of the invention；At the same time, for those skilled in the art, foundation
Thought of the invention, there will be changes in the specific implementation manner and application range.In conclusion the content of the present specification is not
It is interpreted as limitation of the present invention.
Claims (8)
1. a kind of cerebral apoplexy Method of EEG signals classification, which is characterized in that the described method includes:
Obtain cerebral apoplexy EEG signals to be sorted；
The cerebral apoplexy EEG signals to be sorted are pre-processed；
Layered shaping is carried out to pretreated cerebral apoplexy EEG signals to be sorted；
The hierarchical fuzzy entropy characteristic value of cerebral apoplexy EEG signals to be sorted after extracting layering；
Obtain disaggregated model；
According to the hierarchical fuzzy entropy characteristic value, by the disaggregated model, to the cerebral apoplexy EEG signals to be sorted into
Row classification.
2. cerebral apoplexy Method of EEG signals classification according to claim 1, which is characterized in that described to described to be sorted
Cerebral apoplexy EEG signals are pre-processed, and are specifically included:
Low-pass filtering is carried out to the cerebral apoplexy EEG signals to be sorted；
Segment processing is first carried out to filtered cerebral apoplexy EEG signals.
3. cerebral apoplexy Method of EEG signals classification according to claim 1, which is characterized in that in acquisition classification mould
Type, before further include:
Obtain supporting vector machine model and training sample；The training sample is cerebral apoplexy brain electricity sample signal, the brain soldier
Midbrain electricity sample signal includes cerebral infarction sample signal and cerebral hemorrhage sample signal；
The training sample is pre-processed；
Layered shaping is carried out to pretreated training sample；
The fuzzy entropy characteristic value of training sample after extracting layered shaping；
By the sample hierarchical fuzzy entropy characteristic value training supporting vector machine model, disaggregated model is obtained.
4. cerebral apoplexy Method of EEG signals classification according to claim 1, which is characterized in that further include: pass through principal component
Analytic approach carries out dimensionality reduction to hierarchical fuzzy entropy characteristic value.
5. a kind of cerebral apoplexy eeg signal classification system, which is characterized in that the system comprises:
First obtains module, for obtaining cerebral apoplexy EEG signals to be sorted；
First preprocessing module, for being pre-processed to the cerebral apoplexy EEG signals to be sorted；
First layer module extracts mould for carrying out layered shaping first to pretreated cerebral apoplexy EEG signals to be sorted
Block, for extracting the hierarchical fuzzy entropy characteristic value of the cerebral apoplexy EEG signals to be sorted after being layered；
Model obtains module, for obtaining disaggregated model；
Categorization module is used for according to the hierarchical fuzzy entropy characteristic value, by the disaggregated model, to the brain soldier to be sorted
Middle EEG signals are classified.
6. cerebral apoplexy eeg signal classification system according to claim 5, which is characterized in that the first pretreatment mould
Block specifically includes:
Filter unit, for carrying out low-pass filtering to the cerebral apoplexy EEG signals to be sorted；
Segmenting unit, for first carrying out segment processing to filtered cerebral apoplexy EEG signals.
7. cerebral apoplexy eeg signal classification system according to claim 5, which is characterized in that the categorizing system is also wrapped
It includes:
Second obtains module, for obtaining supporting vector machine model and training sample；The training sample is cerebral apoplexy brain electricity
Sample signal, the cerebral apoplexy brain electricity sample signal include cerebral infarction sample signal and cerebral hemorrhage sample signal；
Second preprocessing module, for being pre-processed to the training sample；
Second hierarchical block, for carrying out layered shaping to pretreated training sample
Second extraction module, for extracting the fuzzy entropy characteristic value of the training sample after layered shaping；
Training module, for obtaining disaggregated model by the sample hierarchical fuzzy entropy characteristic value training supporting vector machine model.
8. cerebral apoplexy eeg signal classification system according to claim 5, which is characterized in that the categorizing system is also wrapped
It includes:
Dimensionality reduction module, for carrying out dimensionality reduction to hierarchical fuzzy entropy characteristic value by Principal Component Analysis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910278823.3A CN109805928A (en) | 2019-04-09 | 2019-04-09 | A kind of cerebral apoplexy Method of EEG signals classification and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910278823.3A CN109805928A (en) | 2019-04-09 | 2019-04-09 | A kind of cerebral apoplexy Method of EEG signals classification and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109805928A true CN109805928A (en) | 2019-05-28 |
Family
ID=66611710
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910278823.3A Pending CN109805928A (en) | 2019-04-09 | 2019-04-09 | A kind of cerebral apoplexy Method of EEG signals classification and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109805928A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110991364A (en) * | 2019-12-09 | 2020-04-10 | 四川大学 | Electroencephalogram signal classification method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103858433A (en) * | 2011-08-25 | 2014-06-11 | 汤姆逊许可公司 | Hierarchical entropy encoding and decoding |
| CN105956623A (en) * | 2016-05-04 | 2016-09-21 | 太原理工大学 | Epilepsy electroencephalogram signal classification method based on fuzzy entropy |
| CN107622260A (en) * | 2017-10-26 | 2018-01-23 | 杭州电子科技大学 | Lower limb gait phase identification method based on multi-source bio signal |
| CN107837083A (en) * | 2017-10-31 | 2018-03-27 | 太原理工大学 | J ripple automatic testing methods based on least square method supporting vector machine |
| CN108968915A (en) * | 2018-06-12 | 2018-12-11 | 山东大学 | Sleep state classification method and system based on entropy feature and support vector machines |
-
2019
- 2019-04-09 CN CN201910278823.3A patent/CN109805928A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103858433A (en) * | 2011-08-25 | 2014-06-11 | 汤姆逊许可公司 | Hierarchical entropy encoding and decoding |
| CN105956623A (en) * | 2016-05-04 | 2016-09-21 | 太原理工大学 | Epilepsy electroencephalogram signal classification method based on fuzzy entropy |
| CN107622260A (en) * | 2017-10-26 | 2018-01-23 | 杭州电子科技大学 | Lower limb gait phase identification method based on multi-source bio signal |
| CN107837083A (en) * | 2017-10-31 | 2018-03-27 | 太原理工大学 | J ripple automatic testing methods based on least square method supporting vector machine |
| CN108968915A (en) * | 2018-06-12 | 2018-12-11 | 山东大学 | Sleep state classification method and system based on entropy feature and support vector machines |
Non-Patent Citations (3)
| Title |
|---|
| ONNY SATYAWATI ET AL: "Feature Selection for The Classification of Clinical Data of Stroke Patients", 《2017 20TH INTERNATIONAL COFERENCE ON ELECTRICAL MACHINES AND SYSTEMS(ICEMS)》 * |
| 王灿 等: "面向特征融合的脑卒中脑电信号分类方法", 《计算机工程与应用》 * |
| 邹晓阳 等: "基于多尺度模糊熵的动作表面肌电信号模式识别", 《生物医学工程学杂志》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110991364A (en) * | 2019-12-09 | 2020-04-10 | 四川大学 | Electroencephalogram signal classification method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106886792B (en) | Electroencephalogram emotion recognition method for constructing multi-classifier fusion model based on layering mechanism | |
| Bajaj et al. | A hybrid method based on time–frequency images for classification of alcohol and control EEG signals | |
| CN102779234B (en) | A kind of cardiogram classification processing method and device | |
| CN110432898A (en) | A kind of epileptic attack eeg signal classification system based on Nonlinear Dynamical Characteristics | |
| CN110236533A (en) | Epileptic seizure prediction method based on the study of more deep neural network migration features | |
| CN106108894A (en) | A kind of emotion electroencephalogramrecognition recognition method improving Emotion identification model time robustness | |
| CN104887224A (en) | Epileptic feature extraction and automatic identification method based on electroencephalogram signal | |
| CN109934089A (en) | Multistage epileptic EEG Signal automatic identifying method based on supervision gradient lifter | |
| CN108511055A (en) | Ventricular premature beat identifying system and method based on Multiple Classifier Fusion and diagnostic rule | |
| CN110141226A (en) | Sleep mode automatically method, apparatus, computer equipment and computer storage medium by stages | |
| Li et al. | Analysis and classification of EEG signals using a hybrid clustering technique | |
| CN110321783A (en) | A kind of MEG spike detection method and system based on 1D convolutional neural networks | |
| CN104166548A (en) | Deep learning method based on motor imagery electroencephalogram data | |
| CN108256453A (en) | A kind of method based on one-dimensional ECG signal extraction two dimension CNN features | |
| CN113951903B (en) | High-speed railway dispatcher overload state identification method based on electroencephalogram data determination | |
| CN112674782B (en) | Device and method for detecting epileptic-like electrical activity of epileptic during inter-seizure period | |
| Zeng et al. | Detecting seizures from EEG signals using the entropy of visibility heights of hierarchical neighbors | |
| CN107301409A (en) | Learn the system and method for processing electrocardiogram based on Wrapper feature selectings Bagging | |
| CN113768519A (en) | Method for analyzing consciousness level of patient based on deep learning and resting state electroencephalogram data | |
| Cai et al. | Brainsleepnet: Learning multivariate eeg representation for automatic sleep staging | |
| Zheng et al. | EEG emotion recognition based on hierarchy graph convolution network | |
| CN109805928A (en) | A kind of cerebral apoplexy Method of EEG signals classification and system | |
| Liu et al. | Extracting multi-scale and salient features by MSE based U-structure and CBAM for sleep staging | |
| CN106963374A (en) | A kind of brain electro-detection method and device based on S-transformation and deep belief network | |
| CN116807479B (en) | Driving attention detection method based on multi-mode deep neural network |
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: 20190528 |
|
| RJ01 | Rejection of invention patent application after publication |