CN104068849A - Method for automatically identifying and extracting K complex waves in sleep brain waves - Google Patents
Method for automatically identifying and extracting K complex waves in sleep brain waves Download PDFInfo
- Publication number
- CN104068849A CN104068849A CN201410312709.5A CN201410312709A CN104068849A CN 104068849 A CN104068849 A CN 104068849A CN 201410312709 A CN201410312709 A CN 201410312709A CN 104068849 A CN104068849 A CN 104068849A
- Authority
- CN
- China
- Prior art keywords
- sequence
- wave
- signal
- complex
- teager energy
- 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.)
- Granted
Links
Landscapes
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
A method for automatically identifying and extracting K complex waves in sleep brain waves comprises the following steps of performing wavelet decomposition and reconstruction on brain wave signals; performing Teager energy operator calculation on reconstructed data and obtaining absolute values; smoothing and performing 0/1 coarse graining on an obtained Teager energy curve; performing threshold detection on the data which is performed on coarse graining; performing morphology detection on reconstructed signals satisfying the threshold values, enabling the signals at positions which satisfying a morphology condition to be the K complex waves and recording starting and final positions and wave crest and wave trough values and positions. The method for automatically identifying and extracting the K complex waves in the sleep brain waves has the advantages of analyzing the signals which is performed on the wavelet decomposition and the reconstruction through the Teager energy operator, extracting an absolute value sequence of the Teager energy operator and performing smoothness and coarse graining processes, being easy to achieve and high in anti-noise capacity, accurately confirming K complex wave positions and the wave crest and trough values and positions and establishing foundation for identification of a non-rem second period in sleep stage and research of the K complex waves.
Description
Technical field
The present invention relates to EEG signals technical field, particularly a kind of method of K complex wave in automatic identification and extraction sleep cerebral electricity.
Background technology
Sufficient sleep is three health standards that international community generally acknowledges, but along with scientific and technical progress, the quickening of rhythm of life and from the continuous increase of each side pressure, the mankind are more and more suffering the torment of various sleep associated with disease, such as: insomnia, drowsiness etc.The physiological health that the decline of sleep quality not only has influence on people also can endanger people's mental health, because the decline of sleep quality is accompanied by the generation of some mental illness toward contact, such as: anxiety neurosis, depression, chronic pain, fears the symptoms such as insomnia.Except mental sickness, the relation of other diseases and sleep has also obtained research widely, by the detection of sleep electroencephalogram, can provide for epilepsy, cerebral thrombosis, cerebral anoxia, central nervous system disease and disturbance of intelligence the information of the aspects such as diagnosis and treatment.Therefore observation and the monitoring of sleep are had great importance.
The judgement of sleeper effect not only will see the length of the length of one's sleep, the more important thing is to be decided by Depth of sleep.Therefore sleep stage is to research sleep important in inhibiting.In the world sleep is divided into two states at present: nonrapid eye movements (NREM) phase and rapid eye movement phase, the nonrapid eye movements (NREM) phase is divided into again one, two, three and the fourth phase, and its Depth of sleep is deepened successively.It is exactly in electroencephalogram, to occur K complex wave that sleep enters the most important flag sign of nonrapid eye movements (NREM) second phase.K complex wave is the two-phase slow wave of amplitude, in morphology, requires first to have a forward wave, a negative wave and then, average duration is 0.5-1.5s, peak-to-peak value at 100v between 400v.The identification of K complex wave not only provides important differentiation feature for sleep stage, to important in inhibitings such as the physiological mechanism of exploration sleep, sleep disorder, cognitive activities, in clinical sleep disorder disease detection and treatment, has good application prospect.
The method of traditional detection K complex wave mainly relies on manually, not only waste time and energy and also subjectivity strong; Along with the research to K complex wave, engendered the method for a lot of K of identification automatically complex waves, mainly contain following several: morphologic detection, time frequency analysis and extraction feature are passed through neural network classification.Morphologic detection is easily subject to the interference of noise, and neural network classification can not judge the position that K complex wave occurs exactly.Nearly 2 years, external researcher proposed first to adopt digital filtering to detect the K complex wave in brain electricity in conjunction with Teager energy operator method, and recycling wavelet decomposition reconstruct and morphological method detect K complex wave, by both results, contrast, and finally extract K complex wave.
But, because Teager energy operator is a kind of method that is applicable to simple component signal analysis, be easily subject to noise, interference effect; And digital filtering technique can not be removed noise, the interference of low-frequency range well completely, thereby cause filtering for the identification of K complex wave in clinical practice signal, to have larger error in conjunction with the method for energy operator, finally cause whole detection poor effect, versatility, the robustness of algorithm are poor.
Summary of the invention
In order to overcome the defect of above-mentioned prior art; the object of the present invention is to provide the method for K complex wave in a kind of automatic identification and extraction sleep cerebral electricity; first adopt wavelet decomposition and reconstructing method to extract the signal of K complex wave place corresponding band; then to reconstruction signal, adopt Teager energy operator to analyze and extract Teager energy operator sequence; calculate the absolute value sequence of this sequence, and to its carry out smoothly, coarse processes.These measures have guaranteed, by the precision of the identification of Teager energy operator and extraction K complex wave, to have removed the harmful effect of noise for the method, have improved anti-noise ability and the robustness of algorithm.
In order to achieve the above object, the present invention is achieved through the following technical solutions:
A method for K complex wave in automatic identification and extraction sleep cerebral electricity, specific implementation step is as follows:
Step 1, read sleep cerebral electricity signal x (t), to x (t) segmentation, segment length is 30s, to the data of each section, utilize db5 small echo to carry out the wavelet decomposition of eight layers, main distribution according to K complex wave, select the low frequency coefficient of the 7th layer, 8 layers to be reconstructed, obtain EEG signals y (t) after pretreatment;
Step 2, the maximum max of EEG signals y (t) after pretreatment (y (t)) and minima min (y (t)) are contrasted, when if the absolute value of the difference of maximum and minima is greater than 100 μ v, EEG signals y (t) after pretreatment is carried out the calculating of Teager energy operator (TEO) and gets its absolute value, obtain the absolute value sequence z (n) of Teager energy operator (TEO), if do not meet above-mentioned condition, do not process, carry out the processing of next segment signal;
Step 3, the absolute value sequence z (n) of Teager energy operator (TEO) is carried out to smoothing processing, method is averaged for adding rectangular window, and window length is 125, and step-length is 1, obtains new sequence o (n);
Step 4, the sequence o (n) after level and smooth is carried out to coarse be converted into 0/1 sequence p (n);
Continuous 1 persistent period in step 5, detection 0/1 sequence p (n), if the persistent period is greater than 0.5s, record the position that its initial sum stops, as certificate, go to search the corresponding position of reconstruction signal, then the reconstruction signal of this position is carried out to morphologic detection, meet morphologic detection condition the signal of this position be K complex wave, and record the value of starting point and final position and Wave crest and wave trough and its position.
In described step 2, EEG signals y (t) after pretreatment is carried out to the calculating of Teager energy operator (TEO), is specially:
Teager energy operator is defined as:
Wherein, x (n) represents after pretreatment n sample value signal in EEG signals y (t),
represent the output sequence of Teager energy operator (TEO).
In described step 4, the formula that coarse grain turns to 0/1 sequence p (n) is:
Wherein o (n) is institute's calling sequence after smoothing processing; the threshold value of threshold for setting; it is defined as: threshold=1.5 * (mean (o (n))+(std (o (n)))); mean (o (n)) is the meansigma methods of o (n); std (o (n)) is the standard deviation of o (n), and p (n) is the 0/1 new sequence of the rear gained of coarse conversion.
In described step 5, the detailed process of morphologic detection is: the peak-to-peak value of first asking the corresponding position of reconstruction signal, be the peak-to-peak value of the reconstruction signal that continuous 1 original position and final position are corresponding, whether detected peaks peak value meets the condition that is greater than 100v and is less than 400v; Next detects whether satisfied 1/4 of the total duration that is greater than of persistent period of forward and negative wave; Whether the slope between last detected peaks peak value meets is greater than 200v/s; If above-mentioned condition all meets, the signal of this time period is K complex wave.
The method for automatic identification and extraction sleep cerebral electricity K complex wave proposing in the present invention; first adopt wavelet decomposition and reconstructing method to extract the signal of K complex wave place corresponding band; then to reconstruction signal, adopt Teager energy operator to analyze and extract Teager energy operator sequence; calculate the absolute value of this sequence; and carry out position level and smooth, that coarse processes to detect K complex wave; the wave character again this position being carried out in time domain detects, and judges whether to meet the morphology requirement of K complex wave.
The present invention can identify and extract the K complex wave in sleep cerebral electricity well automatically, algorithm is easy, capacity of resisting disturbance is strong, the position that simultaneously can mark more exactly K complex wave with and value and the position of Wave crest and wave trough, in addition also for the identification of 2 phases of nonrapid eye movements (NREM) and the research of K complex wave are laid a good foundation.
Accompanying drawing explanation
Fig. 1 is algorithm overview flow chart of the present invention.
Fig. 2 is the design sketch of the signal after original signal and wavelet reconstruction, and wherein Fig. 2 A is the design sketch of original signal, and Fig. 2 B is the design sketch of the signal after wavelet reconstruction.
Fig. 3 is the design sketch of the absolute value sequence of TEO.
Fig. 4 is the design sketch of the TEO sequence after level and smooth.
Fig. 5 is the design sketch of institute's calling sequence after 0/1 conversion.
Fig. 6 is the design sketch of the result of K complex wave detection.
The specific embodiment
Below in conjunction with accompanying drawing, the present invention is further described.
Automatic identification and extract the method for K complex wave in sleep cerebral electricity, the concrete implementation step of its realization is with reference to Fig. 1:
Step 1, read sleep cerebral electricity and to its segmentation, because sleep stage generally be take 30s as first phase, therefore the segment length of segmentation is 30s, according to the frequency domain feature of K complex wave: its frequency is generally at 0.5-2Hz, therefore select db5 small echo to carry out the wavelet decomposition of eight layers, select the low frequency coefficient of the 7th, 8 layers to be reconstructed, Fig. 2 is exactly the signal after original signal and wavelet reconstruction.
The amplitude of step 2, K complex wave is very large, and generally at 100-400v, therefore the absolute value to the difference of reconstruction signal maximizing and minima, if while being greater than 100v, carrying out the calculating of TEO and get its absolute value it, obtains the absolute value sequence of TEO; Otherwise this segment signal does not comprise K complex wave, does not process, carry out the processing of next segment signal;
Wherein TEO is defined as:
Wherein, x (n) represents after pretreatment n sample value signal in EEG signals y (t),
represent the output sequence of Teager energy operator (TEO), the absolute value sequence that Fig. 3 is TEO.
Step 3, in order to detect better the situation of change of TEO, need to carry out smoothing processing to TEO sequence, adding rectangular window averages, window length is 125, step-length is 1, in order to obtain the new sequence with the same length of TEO sequence, before windowing, need after former data, mend 124 values, mend to such an extent that value is the amplitude of last point, Fig. 4 be the design sketch of the TEO sequence after smoothly.
Step 4, the TEO sequence coarse grain after level and smooth is turned to 0/1 sequence p (n), the standard of coarse is:
Wherein o (n) is former sequence; the threshold value of threshold for setting; it is defined as: threshold=1.5 * (mean (o (n))+(std (o (n)))); mean (o (n)) is the meansigma methods of o (n); std (o (n)) is the standard deviation of o (n), and p (n) is the 0/1 new sequence after coarse conversion.The setting of this threshold value is that many experiments checking obtains, and Fig. 5 is the result of 0/1 conversion.
Step 5, because persistent period of K complex wave is generally 0.5s-1.5s, therefore detect in 0/1 sequence for continuous 1 persistent period, if the persistent period is greater than 0.5s, this segment data is likely K complex wave, records the position that its initial sum stops; In order to judge whether more exactly it is K complex wave, to carry out morphologic detection to the reconstruction signal on correspondence position, and record value and its position of Wave crest and wave trough;
Wherein morphologic detection need to meet three conditions:
A. the peak-to-peak value of this segment signal need be greater than 100v and be less than 400v;
B. the persistent period of forward wave and negative wave need be greater than total duration 1/4th;
C. the slope between peak-to-peak value need be not less than 200 μ v/s.
If meet the condition of morphologic detection, illustrate that this segment signal is K complex wave, and on original signal, give mark out by circle, also use " * " number to mark its Wave crest and wave trough simultaneously; ; If do not meet the condition of this morphologic detection, illustrate that this segment signal is not K complex wave, Fig. 6 is the result that K complex wave detects.
Claims (4)
1. automatically identify and extract a method for K complex wave in sleep cerebral electricity, it is characterized in that, specific implementation step is as follows:
Step 1, read sleep cerebral electricity signal x (t), to x (t) segmentation, segment length is 30s, to the data of each section, utilize db5 small echo to carry out the wavelet decomposition of eight layers, main distribution according to K complex wave, select the low frequency coefficient of the 7th, 8 layers to be reconstructed, obtain EEG signals y (t) after pretreatment;
Step 2, the maximum max of EEG signals y (t) after pretreatment (y (t)) and minima min (y (t)) are contrasted, when if the absolute value of the difference of maximum and minima is greater than 100 μ v, EEG signals y (t) after pretreatment is carried out the calculating of Teager energy operator (TEO) and gets its absolute value, obtain the absolute value sequence z (n) of Teager energy operator (TEO), if do not meet above-mentioned condition, do not process, carry out the processing of next segment signal;
Step 3, the absolute value sequence z (n) of Teager energy operator (TEO) is carried out to smoothing processing, method is averaged for adding rectangular window, and window length is 125, and step-length is 1, obtains new sequence o (n);
Step 4, the sequence o (n) after level and smooth is carried out to coarse be converted into 0/1 sequence p (n);
Continuous 1 persistent period in step 5, detection 0/1 sequence p (n), if the persistent period is greater than 0.5s, record the position that its initial sum stops, as certificate, go to search the corresponding position of reconstruction signal, then the reconstruction signal of this position is carried out to morphologic detection, meet morphologic detection condition the signal of this position be K complex wave, and record the value of starting point and final position and Wave crest and wave trough and its position.
2. the method for K complex wave in a kind of automatic identification according to claim 1 and extraction sleep cerebral electricity, is characterized in that, in described step 2, EEG signals y (t) after pretreatment is carried out to the calculating of Teager energy operator (TEO), is specially:
Teager energy operator is defined as:
Wherein, x (n) represents after pretreatment n sample value signal in EEG signals y (t),
represent the output sequence of Teager energy operator (TEO).
3. the method for K complex wave in a kind of automatic identification according to claim 1 and extraction sleep cerebral electricity, is characterized in that, in described step 4, the formula that coarse grain turns to 0/1 sequence p (n) is:
Wherein o (n) is institute's calling sequence after smoothing processing; the threshold value of threshold for setting; it is defined as: threshold=1.5 * (mean (o (n))+(std (o (n)))); mean (o (n)) is the meansigma methods of o (n); std (o (n)) is the standard deviation of o (n), and p (n) is the 0/1 new sequence of the rear gained of coarse conversion.
4. a kind of automatic identification according to claim 1 and extract the method for K complex wave in sleep cerebral electricity, it is characterized in that, in described step 5, the detailed process of morphologic detection: the peak-to-peak value of first asking the corresponding position of reconstruction signal, be the peak-to-peak value of the reconstruction signal that continuous 1 original position and final position are corresponding, whether detected peaks peak value meets the condition that is greater than 100v and is less than 400v; Next detects whether satisfied 1/4 of the total duration that is greater than of persistent period of forward and negative wave; Whether the slope between last detected peaks peak value meets is greater than 200v/s; If above-mentioned condition all meets, the signal of this time period is K complex wave.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410312709.5A CN104068849B (en) | 2014-07-02 | 2014-07-02 | A kind of automatic identification and extract the method for K complex wave in sleep cerebral electricity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410312709.5A CN104068849B (en) | 2014-07-02 | 2014-07-02 | A kind of automatic identification and extract the method for K complex wave in sleep cerebral electricity |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104068849A true CN104068849A (en) | 2014-10-01 |
| CN104068849B CN104068849B (en) | 2016-04-13 |
Family
ID=51590751
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410312709.5A Active CN104068849B (en) | 2014-07-02 | 2014-07-02 | A kind of automatic identification and extract the method for K complex wave in sleep cerebral electricity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104068849B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106166068A (en) * | 2016-09-21 | 2016-11-30 | 广州视源电子科技股份有限公司 | Method and system for marking electroencephalogram data types in sleep state |
| CN106175673A (en) * | 2016-07-04 | 2016-12-07 | 西安交通大学 | A kind of method of spindle wave in automatic identification and extraction sleep cerebral electricity |
| CN106175698A (en) * | 2016-09-21 | 2016-12-07 | 广州视源电子科技股份有限公司 | Sleep cycle detection device in sleep state analysis |
| CN106175755A (en) * | 2016-09-21 | 2016-12-07 | 广州视源电子科技股份有限公司 | Sleep state detector for sleep state analysis |
| CN106333678A (en) * | 2016-09-21 | 2017-01-18 | 广州视源电子科技股份有限公司 | Method and system for detecting brain waves in sleep state in electroencephalogram signals |
| CN106333674A (en) * | 2016-09-21 | 2017-01-18 | 广州视源电子科技股份有限公司 | Sleep cycle detection method and system in sleep state analysis |
| CN106377250A (en) * | 2016-09-21 | 2017-02-08 | 广州视源电子科技股份有限公司 | Marking equipment for electroencephalogram data types in sleep state |
| CN106377249A (en) * | 2016-09-21 | 2017-02-08 | 广州视源电子科技股份有限公司 | Sleep state detection method and system in sleep state analysis |
| CN110432865A (en) * | 2019-07-24 | 2019-11-12 | 喜临门家具股份有限公司 | A kind of sleep state classification system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5626627A (en) * | 1995-07-27 | 1997-05-06 | Duke University | Electroconvulsive therapy method using ICTAL EEG data as an indicator of ECT seizure adequacy |
| CN102393874A (en) * | 2011-07-01 | 2012-03-28 | 山东大学 | Electroencephalogram spike wave frequency based early-alarming method and device |
| CN103690169A (en) * | 2013-12-31 | 2014-04-02 | 中国科学院深圳先进技术研究院 | Respiration information detection method and system |
-
2014
- 2014-07-02 CN CN201410312709.5A patent/CN104068849B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5626627A (en) * | 1995-07-27 | 1997-05-06 | Duke University | Electroconvulsive therapy method using ICTAL EEG data as an indicator of ECT seizure adequacy |
| CN102393874A (en) * | 2011-07-01 | 2012-03-28 | 山东大学 | Electroencephalogram spike wave frequency based early-alarming method and device |
| CN103690169A (en) * | 2013-12-31 | 2014-04-02 | 中国科学院深圳先进技术研究院 | Respiration information detection method and system |
Non-Patent Citations (6)
| Title |
|---|
| AYKUT ERDAMAR等: "A wavelet and teager energy operator based method for automatic detection of K-complex in sleep EEG", 《EXPERT SYSTEMS WITH APPLICATIONS》, vol. 39, 31 December 2012 (2012-12-31), pages 1284 - 1286 * |
| C.KAMATH: "ECG beat classification using features extracted from Teager energy functions in time and frequency domains", 《IET SIGNAL PROCESSING》, vol. 5, no. 6, 31 December 2011 (2011-12-31), pages 575 - 581, XP006043152, DOI: doi:10.1049/iet-spr.2010.0138 * |
| 宦飞等: "利用小波级数检测睡眠EEG中的K-复合波", 《北京生物医学工程》, vol. 19, no. 1, 31 March 2000 (2000-03-31) * |
| 曾番: "小波包分析与能量算子结合的声目标特征参数", 《探测与控制学报》, vol. 32, no. 6, 31 December 2010 (2010-12-31) * |
| 王义元等: "基于小波变换和Teager能量算子浊音段提取", 《控制工程》, vol. 11, 31 July 2004 (2004-07-31) * |
| 陈金燕等: "基于小波变换和Teager能量算子的癫痫脑电自动分类", 《生物医学工程学进展》, vol. 33, no. 3, 30 September 2012 (2012-09-30) * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106175673B (en) * | 2016-07-04 | 2019-01-18 | 西安交通大学 | A kind of method of automatic identification and spindle wave in extraction sleep cerebral electricity |
| CN106175673A (en) * | 2016-07-04 | 2016-12-07 | 西安交通大学 | A kind of method of spindle wave in automatic identification and extraction sleep cerebral electricity |
| CN106333678A (en) * | 2016-09-21 | 2017-01-18 | 广州视源电子科技股份有限公司 | Method and system for detecting brain waves in sleep state in electroencephalogram signals |
| CN106175698B (en) * | 2016-09-21 | 2019-05-07 | 广州视源电子科技股份有限公司 | Sleep cycle detection device in sleep state analysis |
| CN106166068A (en) * | 2016-09-21 | 2016-11-30 | 广州视源电子科技股份有限公司 | Method and system for marking electroencephalogram data types in sleep state |
| CN106333674A (en) * | 2016-09-21 | 2017-01-18 | 广州视源电子科技股份有限公司 | Sleep cycle detection method and system in sleep state analysis |
| CN106377250A (en) * | 2016-09-21 | 2017-02-08 | 广州视源电子科技股份有限公司 | Marking equipment for electroencephalogram data types in sleep state |
| CN106377249A (en) * | 2016-09-21 | 2017-02-08 | 广州视源电子科技股份有限公司 | Sleep state detection method and system in sleep state analysis |
| CN106175698A (en) * | 2016-09-21 | 2016-12-07 | 广州视源电子科技股份有限公司 | Sleep cycle detection device in sleep state analysis |
| CN106175755A (en) * | 2016-09-21 | 2016-12-07 | 广州视源电子科技股份有限公司 | Sleep state detector for sleep state analysis |
| CN106333678B (en) * | 2016-09-21 | 2019-05-07 | 广州视源电子科技股份有限公司 | Method and system for detecting brain waves in sleep state in electroencephalogram signals |
| CN106175755B (en) * | 2016-09-21 | 2019-07-09 | 广州视源电子科技股份有限公司 | Sleep state detector for sleep state analysis |
| CN106377249B (en) * | 2016-09-21 | 2019-07-09 | 广州视源电子科技股份有限公司 | Sleep state detection method and system in sleep state analysis |
| CN106377250B (en) * | 2016-09-21 | 2019-07-09 | 广州视源电子科技股份有限公司 | Marking equipment for electroencephalogram data types in sleep state |
| CN106333674B (en) * | 2016-09-21 | 2019-07-09 | 广州视源电子科技股份有限公司 | Sleep cycle detection method and system in sleep state analysis |
| CN110432865A (en) * | 2019-07-24 | 2019-11-12 | 喜临门家具股份有限公司 | A kind of sleep state classification system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104068849B (en) | 2016-04-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104068849B (en) | A kind of automatic identification and extract the method for K complex wave in sleep cerebral electricity | |
| Puthankattil et al. | Classification of EEG signals in normal and depression conditions by ANN using RWE and signal entropy | |
| CN104367316B (en) | Denoising of ECG Signal based on morphologic filtering Yu lifting wavelet transform | |
| Liu et al. | Exploring the time–frequency content of high frequency oscillations for automated identification of seizure onset zone in epilepsy | |
| CN105411565B (en) | Heart rate variability tagsort method based on broad sense multi-scale wavelet entropy | |
| Koolen et al. | Line length as a robust method to detect high-activity events: automated burst detection in premature EEG recordings | |
| Lopes et al. | Automatic electroencephalogram artifact removal using deep convolutional neural networks | |
| Lai et al. | Automated detection of high frequency oscillations in intracranial EEG using the combination of short-time energy and convolutional neural networks | |
| CN101259016A (en) | Method for real time automatically detecting epileptic character wave | |
| CN105320969A (en) | A heart rate variability feature classification method based on multi-scale Renyi entropy | |
| Wang et al. | An emotional analysis method based on heart rate variability | |
| Yuan et al. | Automatic seizure detection using diffusion distance and BLDA in intracranial EEG | |
| Chaibi et al. | A comparaison of methods for detection of high frequency oscillations (HFOs) in human intacerberal EEG recordings | |
| Zhang et al. | Automatic detection of interictal epileptiform discharges based on time-series sequence merging method | |
| Yadav et al. | Morphology-based automatic seizure detector for intracerebral EEG recordings | |
| CN103761424A (en) | Electromyography signal noise reducing and aliasing removing method based on second-generation wavelets and ICA (independent component analysis) | |
| CN113768519A (en) | Method for analyzing consciousness level of patient based on deep learning and resting state electroencephalogram data | |
| Wu et al. | EMGdi signal enhancement based on ICA decomposition and wavelet transform | |
| Liu et al. | Systematic methods for fetal electrocardiographic analysis: Determining the fetal heart rate, RR interval and QT interval | |
| Ren et al. | MUAP extraction and classification based on wavelet transform and ICA for EMG decomposition | |
| Bahador et al. | Multimodal spatio-temporal-spectral fusion for deep learning applications in physiological time series processing: A case study in monitoring the depth of anesthesia | |
| Karimui et al. | Cardiac arrhythmia classification using the phase space sorted by Poincare sections | |
| Ru et al. | Epilepsy detection based on variational mode decomposition and improved sample entropy | |
| Jibon et al. | Epileptic seizure detection from electroencephalogram (EEG) signals using linear graph convolutional network and DenseNet based hybrid framework | |
| Yang et al. | Depression detection based on analysis of EEG signals in multi brain regions |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |