CN1954772A - Self-reference, self-correlation and self-adaptive interference balancing method of evoked potential single extracted - Google Patents
Self-reference, self-correlation and self-adaptive interference balancing method of evoked potential single extracted Download PDFInfo
- Publication number
- CN1954772A CN1954772A CNA2006100545066A CN200610054506A CN1954772A CN 1954772 A CN1954772 A CN 1954772A CN A2006100545066 A CNA2006100545066 A CN A2006100545066A CN 200610054506 A CN200610054506 A CN 200610054506A CN 1954772 A CN1954772 A CN 1954772A
- Authority
- CN
- China
- Prior art keywords
- self
- signal
- reference signal
- generated reference
- correlation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
A self-reference, self-correlative and self-adaptive method for canceling the interference in extracting the evoked potential (EP) includes such steps as using a single electrode to take the self-reference signals for a long time, inputting the self-reference signals and comprehensive signals to the amplifier of electrobiologic examine system, amplifying, A/D conversion to obtain digital signals, and processing by special software (including self-reference analysis, calculating cancel factor W and canceling interference) to obtain EP, and calculating EP parameters.
Description
Technical field
The present invention relates to a kind of armarium technology and biomedical engineering technology, design a kind of self-reference, self-correlation and self-adaptive interference cancellation method that the current potential single extracts of bringing out especially.
Background technology
Bring out current potential (EP:Evoked potential) and be called evoked brain potential position (EBP:Evoked brainpotential) again, this is for spontaneous brain electricity position (SBP:Spontaneous brain potential).Bringing out current potential comprises: (1) somatosensory evoked potential, (2) visual evoked potential, (3) auditory evoked potential, (4) cognitive potential, (5) OEP, (6) GEP.Bring out current potential and can feel system disorders such as (look, listen and body sense) in order to assisted diagnosis, research is grown and is aged.Existing research is pointed out, before Histological change, dull-witted (dementia, as AD:Alzheimer ' s disease) the existing EP change of patient, comprise prolongation of latency, amplitude reduces (as p200, p300, N200), a kind of method for early warning of pointing out this method to can be used as senile nervous system disease is so that intervene in early days.But existing method can not reflect the parameter variation between stimulation.The research of electric physiological performance type and genotypic relation simultaneously, be the problem of exploring now, nerve-mental sickness and gene, nerve-mental sickness and gene and the research of getting in touch of bringing out current potential, first meeting clue is also more and more higher to extracting method and the accuracy requirement thereof that brings out current potential.Traditional current potential extracting method that brings out is the superposed average method at present, the basic assumption of superposition average theory: (i) amplitude and the incubation period of the EP of each time stimulation generation are constant; (ii) background signal comprises that spontaneous brain electricity and 50Hz disturb etc. that to be all average be 0 noise.This theoretical defective is: (i) clinical practice proves, each time stimulates amplitude and the incubation period of the EP that produces all may change, especially patient; (ii) background signal comprises that it is not that average is 0 noise that spontaneous brain electricity (S-EEG) and 50Hz disturb etc.Because it is to determine signal that 50Hz disturbs, S-EEG has α, β, δ etc. clinical in it specific waveforms of diagnosing, neither white noise.Like this, limited number of time superposition averaged result is exactly the superposition average result that the superposition average result of EP adds 50Hz signal and S-EEG, because S-EEG is bigger one more than the magnitude than EP, so the result is had bigger influence, the particularly result dependency to the superposition average time.If EP (t), BG (t) represent to bring out current potential and whole background signal (non-echo signal) respectively, then N superposition averaged result is:
∑S(t)/N＝∑EP(t)/N+∑BG(t)/N
S in the formula (t)=EP (t)+BG (t).The special N=N that works as
iWhen inferior BG (t) be able to eliminate, N
iTo cause for+1 time huge background signal BG to flood EP.In view of this, traditional accuracy that brings out current potential extracting method gained result is not high, more and more can not satisfy the development need of modern medicine.
Summary of the invention
At present technical problem, the present invention is according to bringing out the particularity that current potential extracts environment, a kind of more advanced self-reference, self correlation, adaptive interference cancellation method (The Auto-reference, Auto-correlation and Auto-adaptation Interference Cancellation Method (AAA-ICM) for ExtractingEvoked Potential by Single Trial) of bringing out the extraction of current potential single more accurately that is fit to this environment is provided
Concrete technical scheme is as follows:
Bring out self-reference, self correlation, adaptive interference cancellation method that the current potential single extracts, may further comprise the steps successively: (1) obtains self-generated reference signal person under inspection dorsal position, according to the different choice body surface position of bringing out current potential of being extracted, install each electrode pair in corresponding position, adopt the electrophysiologic study system to obtain spontaneous brain electricity signal more than at least one minute as self-generated reference signal;
(2) obtain integrated signal when obtaining the self-generated reference signal end, give stimulus signal automatically, obtain integrated signal, guarantee that the sampling of self-generated reference signal and integrated signal is carried out continuously with the electrode that obtains self-generated reference signal by computer-controlled system;
(3) ask the slip self correlation to establish
EP(n)＝S(n)-REF(n) (1)
In the formula, current potential is brought out in EP (n) expression, the integrated signal that S (n) expression step (2) obtains, for bringing out current potential+ambient interferences, REF (n) is the self-generated reference signal that step (1) obtains, method and measure guarantees that the length of REF (n) is more than at least 20 times of S (n), asks the slip self correlation of S (n) and REF (n):
R(m)＝REF(n)·S(n) (2)
Obtain autocorrelation coefficient series R (m); In the wherein said formula (2), " " is relevant operator;
(4) ask maximum correlation coefficient from R (m), to obtain maximum coefficient R
Max(m), m is exactly candidate's an optimum self-generated reference signal section starting point, from then on begin to get with the isometric signal segment of integrated signal as optimum self-generated reference signal, use REF
M(n) expression;
(5) asking W is that the self adaptation regulatory factor W that self adaptation is obtained with adaptive approach regulates, and algorithm is as follows:
If
EP(n)＝S(n)-WREF
M(n) (3)
W is a regulatory factor in the formula (3), regulates W and makes EP (n) minimum, asks W by the minimum power criterion, order:
D[∑ EP
2(n)]/dw=d{ ∑ [S (n)-WREF
M(n)]
2}/dw=0 (4) gets W=∑ S (n) REF
M(n)/∑ [REF
M(n)]
2(5)
(6) interference cancellation is after by formula (5) obtain W, and by formula (3) formula realizes offseting, and tries to achieve the best and offsets EP (n) under the condition.
The sample frequency of self-generated reference signal is identical with the sample frequency of integrated signal in the described step (1).
The sample frequency of described self-generated reference signal and the sample frequency of integrated signal are 1000～2000 hertz.
The time that is obtained from the carbuncle in the occipital region signal of telecommunication in the described step (1) is 1～2 minute, and the time of obtaining integrated signal is 0.25～1 second.
Generally speaking, the electric potential signal that brings out that is obtained by step (6) still contains some random noises, because random noise generally can not eliminate fully, in order further to eliminate this random noise, can use various filtering techniques.
In step (1), (2) of the method is to adopt the electrophysiologic study system, the person under inspection is carried out the extraction of self-generated reference signal and integrated signal, according to the difference of being extracted of bringing out current potential, give mode to the installation site of signal electrode (+) and stimulation and do corresponding adjustment.The amplifier of the self-generated reference signal that obtains and integrated signal input electrophysiologic study system is amplified, carry out A/D conversion becoming digital signal again, utilize then according to the software that formula (1)～write (5) data are handled, comprise from coherence analysis, calculating offsets factor W, interference cancellation brings out current potential EP to obtain, and carries out the EP calculation of parameter at last.
The acquisition of self-generated reference signal is the embodiment of core novelty of the present invention, because method in the past is to use many references or other reference (as Parsa et al.Multireference adaptive noise cancellationapplied to somatosensory evoked potentials.IEEE Trans Biomed Eng.1994Aug; 41 (8): 792-800.).The key idea that obtains self-generated reference signal be obtain contain the comprehensive EEG signals of bringing out current potential before, obtain the self-generated reference signal of one period long period with same electrode.Because this is the signal that obtains with same electrode at same position, thus strong dependency can be guaranteed, thus guaranteed the premium properties that offsets.Do not need like existing superposed average method such, amplitude and the incubation period of supposing each stimulation back EP all are constant, extraction facts have proved but be absorbed in, amplitude and incubation period are to stimulating the EP of number of times dependence, so that provide meticulous information to amplitude and the preclinical variation of the EP under disease and intervention (as the medicine) condition; The present invention proposes the new ideas of interfering signal and echo signal; Here EP is echo signal (objective or target signal), do not need as existing superposed average method, suppose that EP determines signal (amplitude and the incubation period that are EP do not become with the stimulation number of times), and the character of EP is not done the supposition or the restriction of any apriority; Also unlike existing superposed average method, suppose the background signal or interfering signal (the interference signal: comprise background EEG that belong to non-echo signal, 50Hz disturbs, and the random disturbances of environment, myoelectricity) be stochastic signal, because clearly background EEG is not a stochastic signal, because clinical will the diagnosis according to its wave character of determining (as α, β, δ etc.).According to objective analysis to the ambient interferences signal, the ambient interferences signal mainly is spontaneous brain electricity signal (spontaneous electroencephalo-signal), next is that 50Hz disturbs, they are not stochastic signal (or being called white noise), the single of EP is extracted, more they can not be considered as stochastic signal or average and be 0 white noise.Obviously, it is periodic signal that 50Hz disturbs, and background brain electricity meets the characteristics of general medicine signal, belongs to nonlinear properties.So the result who adopts extracting method of the present invention to obtain is more accurate.
Single based on theory and technology of the present invention extracts simulation result: can extract echo signal satisfactorily, signal curve such as accompanying drawing 3.In the accompanying drawing 3, the emulation EP signal that the upper left quarter figure a) extracts for single, lower left quarter figure c) the emulation EP signal that produces for computer, bottom-right graph d) for the reference signal of emulation, top right plot b) is the relevant situation that offsets, light color is emulation reference signal (an ambient interferences signal), and dark color is an integrated signal.The result who offsets produces the picture left above extraction signal a).Relatively the emulation EP signal graph that extracts of single of the present invention a) and the emulation EP signal c that produces of computer), two figure signal powers are more or less the same, the waveform tendency is basic identical, this shows checking, and the current potential that brings out that extracts based on theory and technology single of the present invention is accurately.
The comparison of the superposed average technology that the present invention and prior art adopt:
Extract based on the single of the VEP (visual evoked potential) of the human experimentation of the technology of the present invention and the results are shown in accompanying drawing 4, accompanying drawing 5 is typical consequence curve charts of existing superposition averaging as a comparison.
More as can be seen, the trend of two figure is similar from figure, contains more details but figure single a) extracts the result.Whether these details have physiological significance and which type of physiological significance are arranged, and remain further to be studied.According to most labelling conventions of bringing out current potential, P represents crest, and N represents trough.
The invention has the beneficial effects as follows and refreshed the existing extracting method that brings out current potential, provide stable for clinical, the current potential information of bringing out with variation features between stimulation, for the exploitation of novel evoked potentuial measuring system provides the method basis, for genotype with bring out the phenotypic research of current potential and provide more reliable, have more distinctive EP information, for the early warning of senile nervous system disease provides more reliable, more accurate index.
Description of drawings
Fig. 1 is a BIO2000 examination of visual electro physiology system principle diagram
Fig. 2 is an electrode configuration schematic diagram among the embodiment
Fig. 3 adopts the inventive method computer artificial result figure to comprise that a) single extracts signal (the emulation single extracts EP); B) interference cancellation emulation; C) signal (emulation EP) of computer generation; D) the emulation reference signal of computer generation;
Fig. 4 be adopt the inventive method single extract VEP (visual evoked potential) the result curve figure of human body
Fig. 5 be adopt the superposed average technology extract VEP (visual evoked potential) the result curve figure of human body
The single that the specific embodiment human visual brings out current potential (VEP) extracts
The single extraction step of VEP:
(1) obtains self-generated reference signal
Instrument: examination of visual electro physiology system, the BIO2000 examination of visual electro physiology system that adopts Chongqing shellfish Australia Electronic Instrument, Limited to produce.
System principle diagram as shown in Figure 1.With device initialize, person under inspection's dorsal position is worn protective eye lens, the scalp of anhydrous alcohol cleaning electrode lay down location, and electrode surface is coated with special conductive paste, and two signal electrodes are fastened on the scalp with elastic cord, and reference electrode is fixed with medical proof fabric.Sample frequency: 2kHz; Obtain 2 minutes self-generated reference signal.Electrode configuration schematic diagram such as accompanying drawing 2, the electrode of employing such as bottom electrode: silver/silver chloride circular electrode, diameter 9mm.
Active electrode (signal electrode) anode: the shallow nest place of occipital tuberosity top 2cm;
Active electrode (signal electricity level) negative terminal: place, forehead center;
Reference electrode: mastoid process below behind ear-lobe or the ear.
(2) obtain integrated signal when obtaining the self-generated reference signal end, automatically give stimulus signal by computer-controlled system, the employing single flash operation stimulates, sample frequency: 2kHz, acquisition time is 0.5 second, obtain integrated signal with the electrode that obtains self-generated reference signal, guarantee that the sampling of self-generated reference signal and integrated signal is carried out continuously;
(3) amplifier that the self-generated reference signal that extracts is connected the examination of visual electro physiology system with the electrode of integrated signal carries out the A/D conversion, and amplifier is provided with amplification: 20K; Bandwidth: 0.1～75Hz.The ADC:12 position.
The principle of (4) by formula (1)～(5) is write software, and the function in each step is as follows:
From coherence analysis: the slip of asking integrated signal and reference signal is asked for the maximum coherence coefficient from relevant;
Obtain the reference signal section of maximum coherence: REFmax;
Calculating offsets factor W: calculate W according to minimum power (least square) principle;
Interference cancellation: deduct W * REFmax with integrated signal and obtain single extraction signal;
EP calculation of parameter: incubation period and the amplitude of calculating each ripple.
Obtain to bring out electric potential signal (curve chart) through the single extraction that obtains after the above-mentioned processing as shown in Figure 4.
Claims (4)
1. bring out the self-reference, self-correlation and self-adaptive interference cancellation method that the current potential single extracts, may further comprise the steps successively: (1) obtains self-generated reference signal person under inspection dorsal position, according to the different choice body surface position of bringing out current potential of being extracted, install each electrode pair in corresponding position, adopt the electrophysiologic study system to obtain spontaneous brain electricity signal more than at least one minute as self-generated reference signal;
(2) obtain integrated signal when obtaining the self-generated reference signal end, automatically give stimulus signal by computer-controlled system, obtain integrated signal at same position with the electrode that obtains self-generated reference signal, guarantee that the sampling of self-generated reference signal and integrated signal is carried out continuously;
(3) ask the slip self correlation to establish
EP(n)＝S(n)-REF(n) (1)
In the formula, current potential is brought out in EP (n) expression, the integrated signal that S (n) expression step (2) obtains, for bringing out current potential+ambient interferences, REF (n) is the self-generated reference signal that step (1) obtains, method and measure guarantees that the length of REF (n) is more than at least 20 times of S (n), asks the slip self correlation of S (n) and REF (n):
R(m)＝REF(n)·S(n) (2)
Obtain autocorrelation coefficient series R (m); In the wherein said formula (2), " " is relevant operator;
(4) ask maximum correlation coefficient from R (m), to obtain maximum coefficient R
Max(m), m is exactly candidate's an optimum self-generated reference signal section starting point, from then on begin to get with the isometric signal segment of integrated signal as optimum self-generated reference signal, use REF
M(n) expression;
(5) asking W is self adaptation
The self adaptation regulatory factor W that obtains with adaptive approach regulates, and algorithm is as follows:
If
EP(n)＝S(n)-WREF
M(n) (3)
W is a regulatory factor in the formula (3), regulates W and makes EP (n) minimum, asks W by the minimum power criterion, order:
d[∑EP
2(n)]/dw＝d{∑[S(n)-WREF
M(n)]
2}/dw＝0 (4)
Get W=∑ S (n) REF
M(n)/∑ [REF
M(n)]
2(5)
(6) interference cancellation
After by formula (5) obtained W, by formula (3) formula realized offseting, and tried to achieve the best and offseted EP (n) under the condition.
2. self-reference, self-correlation and self-adaptive interference cancellation method of bringing out the extraction of current potential single as claimed in claim 1, the sample frequency of self-generated reference signal is identical with the sample frequency of integrated signal in the step (1).
3. self-reference, self-correlation and self-adaptive interference cancellation method of bringing out the extraction of current potential single as claimed in claim 2, the sample frequency of self-generated reference signal and the sample frequency of integrated signal are 1000～2000 hertz in the step (1).
4. self-reference, self-correlation and self-adaptive interference cancellation method of bringing out the extraction of current potential single as claimed in claim 1, the time that is obtained from the carbuncle in the occipital region signal of telecommunication in the step (1) is 1～2 minute, the time of obtaining integrated signal is 0.25～1 second.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2006100545066A CN1954772A (en) | 2006-10-17 | 2006-10-17 | Self-reference, self-correlation and self-adaptive interference balancing method of evoked potential single extracted |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2006100545066A CN1954772A (en) | 2006-10-17 | 2006-10-17 | Self-reference, self-correlation and self-adaptive interference balancing method of evoked potential single extracted |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1954772A true CN1954772A (en) | 2007-05-02 |
Family
ID=38062464
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2006100545066A Pending CN1954772A (en) | 2006-10-17 | 2006-10-17 | Self-reference, self-correlation and self-adaptive interference balancing method of evoked potential single extracted |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1954772A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101716074B (en) * | 2008-10-09 | 2011-09-14 | 南方医科大学 | Evoked potential recorder based on time characteristic indicators |
| CN102631192A (en) * | 2012-04-24 | 2012-08-15 | 南方医科大学 | Deconvolution method for extracting evoked potential at high stimulation ratio |
| CN105962936A (en) * | 2016-05-17 | 2016-09-28 | 浙江大学 | Isoluminance sinusoidal grating color visual evoked potential diagnosis system |
| CN108309292A (en) * | 2018-04-09 | 2018-07-24 | 中国医学科学院生物医学工程研究所 | A kind of rapid extracting method of Evoked ptential signal |
| CN108882888A (en) * | 2016-03-08 | 2018-11-23 | 克里斯托夫·古格 | Device and method for the electro photoluminescence to subject |
| CN110840454A (en) * | 2019-12-21 | 2020-02-28 | 深圳市杰纳瑞医疗仪器股份有限公司 | Electroencephalogram signal acquisition device and method |
| CN113662564A (en) * | 2021-09-24 | 2021-11-19 | 电子科技大学 | Event evoked potential extraction device based on eigenmode function coherent analysis |
-
2006
- 2006-10-17 CN CNA2006100545066A patent/CN1954772A/en active Pending
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101716074B (en) * | 2008-10-09 | 2011-09-14 | 南方医科大学 | Evoked potential recorder based on time characteristic indicators |
| CN102631192A (en) * | 2012-04-24 | 2012-08-15 | 南方医科大学 | Deconvolution method for extracting evoked potential at high stimulation ratio |
| CN108882888A (en) * | 2016-03-08 | 2018-11-23 | 克里斯托夫·古格 | Device and method for the electro photoluminescence to subject |
| CN108882888B (en) * | 2016-03-08 | 2022-11-22 | 克里斯托夫·古格 | Apparatus and method for electrical stimulation of a subject |
| CN105962936A (en) * | 2016-05-17 | 2016-09-28 | 浙江大学 | Isoluminance sinusoidal grating color visual evoked potential diagnosis system |
| CN108309292A (en) * | 2018-04-09 | 2018-07-24 | 中国医学科学院生物医学工程研究所 | A kind of rapid extracting method of Evoked ptential signal |
| CN110840454A (en) * | 2019-12-21 | 2020-02-28 | 深圳市杰纳瑞医疗仪器股份有限公司 | Electroencephalogram signal acquisition device and method |
| CN113662564A (en) * | 2021-09-24 | 2021-11-19 | 电子科技大学 | Event evoked potential extraction device based on eigenmode function coherent analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1954772A (en) | Self-reference, self-correlation and self-adaptive interference balancing method of evoked potential single extracted | |
| Phinyomark et al. | Fractal analysis features for weak and single-channel upper-limb EMG signals | |
| Merlo et al. | Technical aspects of surface electromyography for clinicians | |
| Usakli | Improvement of EEG signal acquisition: An electrical aspect for state of the art of front end | |
| AU2009294312B2 (en) | Stimulus artifact removal for neuronal recordings | |
| CN110969108B (en) | Limb action recognition method based on autonomic motor imagery electroencephalogram | |
| US20060149139A1 (en) | Apparatus and method for ascertaining and recording electrophysiological signals | |
| Li et al. | ECG denoising method based on an improved VMD algorithm | |
| US10863911B2 (en) | Detecting neuronal action potentials using a convolutive compound action potential model | |
| CN105147282A (en) | Cognitive disorder electroencephalogram recognition system | |
| Smith et al. | Exploiting electrocorticographic spectral characteristics for optimized signal chain design: A 1.08 W analog front end with reduced ADC resolution requirements | |
| CN116671932A (en) | Depression brain electric signal extraction method based on wavelet and self-adaptive filtering | |
| Xu et al. | Digital filter design for peak detection of surface EMG | |
| CN117349598B (en) | Electroencephalogram signal processing method and device, equipment and storage medium | |
| Zhu et al. | A pilot study on the evaluation of normal phonating function based on high-density sEMG topographic maps | |
| CN107510451B (en) | pitch perception ability objective assessment method based on brainstem auditory evoked potentials | |
| CN107374620A (en) | A kind of EEG signals preprocess method based on independent composition analysis algorithm | |
| Tamura et al. | Sleep-stage correlates of hippocampal electroencephalogram in primates | |
| CN112933408B (en) | System and method for collecting surface electromyographic signals and controlling low-frequency electrical stimulation | |
| CN114376585A (en) | Implanted deep electroencephalogram acquisition device | |
| Polyakov et al. | Three-channel Lissajous' trajectory of human middle latency auditory evoked potentials | |
| CN1949140A (en) | Brain-machine interface mouse controlling method | |
| Maby | Technical Requirements for High‐quality EEG Acquisition | |
| Morse et al. | Preferential and non-preferential transmission of formant information by an analogue cochlear implant using noise: the role of the nerve threshold | |
| CN217659897U (en) | Implanted deep electroencephalogram acquisition device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |