CN102579040A - Acupuncture point selecting method for acquiring brain waves - Google Patents

Abstract

The invention discloses an acupuncture point selecting method for acquiring brain waves. A correct acquisition channel is obtained finally. A plurality of brain acupuncture point signals are obtained via electroencephalogram equipment and are analyzed by the aid of a phase space reconstruction method, then embedding dimensions of the various acupuncture point signals are compared, a channel containing more information and a relatively independent channel are used as follow-up researched objects, information of other channels can be reconfigured from a channel with a large embedding dimension, and complexity of acquisition of the brain wave acupuncture point signals is simplified. Real-time analysis of the brain wave signals is strengthened. A mathematical model is simple, a physical significance is clear and definite, and the acupuncture point selecting method is easy to implement.

Description

The acupuncture point system of selection that a kind of brain wave is gathered

Technical field

The present invention relates to the electron medicine field, relate in particular to a kind of eeg signal method of sampling.

Background technology

The brain electricity be cranial nerve cell cerebral cortex and with the bioelectrical activity performance of scalp surface, it can reflect the functional status of brain physiology, pathological condition and brain effectively.Different thought state and emotion changes can both reflect different EEG signals in different cerebral cortex positions, so EEG signals contain abundant useful information.In biomedicine, the brain electricity is used as the effective means of medical diagnosis and disease treatment; In Cognitive Study, the brain electricity is as the important tool of human thinking's origin.That at first the discovery mankind have the brain electrical acti characteristic is the sick scholar Hans of Germanism Berger, and he has delivered the paper of a first piece of writing about human brain electricity in nineteen twenty-nine.

Traditional method is mainly from time domain, frequency spectrum and statistical angle research brain.Time-domain analysis mainly is waveform, the amplitude of research brain electricity, characteristics such as phase place, and frequency-domain analysis method is to analyze the distribution of brain wave on different frequency range.They can describe the Partial Feature of brain electricity, but poor sensitivity, and can't analyze the substitutive characteristics of cerebral activity.Methods such as wavelet transformation, neutral net and nonlinear kinetics also begin to be applied to the analysis of EEG signals in recent years, and they have represented the new direction of electroencephalogramsignal signal analyzing.

Become a very important field with chaology research brain electricity, brain is considered to a nonlinear dynamic system.Divide from the research purpose, mainly contain following two fields: the working mechanism of brain is disclosed in (1), and researching human body is in the Nonlinear Dynamical Characteristics under different physiological statuss, the different cerebral functional status.(2) researching human body is in the variation of nonlinear kinetics under the pathological state, is the clinical foundation of analyzing and diagnosing that provides.The phase space reconfiguration of Whitney delay reconstruct theorem theoretical and Takens provides from the theoretical basis of experimental period sequence reconfiguration system phase space.Based on this theory, Nonlinear Time Series is analyzed and studied, can very effective description goal systems, this also makes the research to the complication system of the unknown become possibility.

When non-linear sequence is carried out phase space reconfiguration; Two major parameters that influence the reconstruct effect are time delay and embedding dimension; Total principle is the nonlinear characteristic that guarantees that the sampling time sequence can be reappeared original system to greatest extent, keeps the information of system as much as possible.Nonlinear Dynamics is the new approaches of research EEG signals, all will play enormous function to deepening the mankind to understanding and disease detection, diagnosis and the treatment of brain.

" Applied Mathematics and Computation " (2009; 207 (1): 63-74) nonlinear dynamic characteristic of EEG signals has been discussed; " IEEE Transactions on Biomedical Engineering " (2011,58 (4): 1084-1093) utilization phase space reconfiguration theoretical research the brain signal situation of change of people when being in different condition.They handle the EEG signals of having gathered; But the collection to EEG signals is but also not mentioned; And we think; The collection of EEG signals is of paramount importance links of EEG Processing, if the signal that collects can comprise maximum quantity of information, will facilitate for subsequent treatment so.This mainly is because brain is not a single node, and the difference of the signal that collects at the brain different parts is very big, how therefrom to select a road optimum signal, makes its can maximum ground characterize the state of brain, comprises maximum quantity of information.Traditional method is generally chosen the signal of head acupoint as candidate signal, but can't make optimized choice to the acupuncture point signal of being gathered.

Summary of the invention

In order to overcome prior art when selecting brain to observe the acupuncture point; The imperfect complexity of bringing with multi-channel measurement of brain wave information; The present invention proposes a kind of new brain wave and observe the method at acupuncture point; Can when guaranteeing that the brain wave information gathering is complete, simplify the complexity that brain wave is gathered, for the plenty of time has been practiced thrift in follow-up brain wave real-time analysis.

The technical solution adopted for the present invention to solve the technical problems may further comprise the steps:

At first, obtain eeg signal, promptly gather the eeg signal of human brain; 32 acupuncture points with brain are acquisition target, make up the acupuncture point passage respectively, through setting-up time length, obtain the eeg signal of 32 acupuncture point passages; The data of 32 acupuncture point passages are stored respectively, obtain the array of 32 equal lengths;

Then, utilize the phase space reconfiguration theory respectively each array to be studied as an independent systematic component, confirm the embedding dimension m and the delay time T of system reconfiguration; At first adopt the FNN method (" Phys.Rev.A ", 1992,45:3403-3411) confirm to embed dimension m, adopt then the AD method (" Physica D ", first extreme point 1994:73:82-98) is as the numerical value of delay time T;

Secondly, the analysis one by one through to each array can obtain m ₁... m _i..m ₃₂, m wherein _iThe embedding dimension of representing the data of i degree passage; Through to embedding the screening of relatively studying passage of dimension: if m _Max＞32 (m _MaxRepresent that the maximum in all passages embeds dimension), increase acquisition target, promptly increase and gather the acupuncture point number, make it equal m _MaxIf, m _Max≤32, then choose and have m _MaxPassage be object of study, keep m simultaneously _i=1 passage; Comprise m _MaxPassage be to comprise the maximum passage of information, m _i=1 passage is the relatively independent passage of E.E.G information;

At last, obtain brain wave reconstruction model information: with m _MaxFor embedding dimension, be time delay with its corresponding τ, the reconfiguration system model:

X(0)＝[x(0)，x(τ)，x(2τ)，L，x((m-1)τ)]

X(1)＝[x(1)，x(τ+1)，x(2τ+1)，L，x((m-1)τ+1)]

...................................

X(n)＝[x(n)，x(n+τ)，x(n+2τ)，L，x(n+(m-1)τ)]

...............................................................

X(m-1)＝[x(m-1)，x(m-1+τ)，x(m-1+2τ)，L，x((m-1)(τ+1))]

This moment X (0) ... all data of .X (m-1) all are from having the maximum dimension m that embeds _MaxPassage, but but can reflect the information of other passage, the information of X (0) expression 1 passage for example, X (1) represent 2 channel informations, X (n) representes the n+1 channel information.Therefore comprise m _MaxPassage comprise except that m _iThe relevant information of other all passages beyond=1 passage, subsequent analysis only need be gathered m _MaxWith the m=1 channel data, need not data acquisition is being carried out at all acupuncture points.

The invention has the beneficial effects as follows: the present invention is from the collection of eeg signal; Utilized the theory of phase space reconfiguration; Embedding dimension through to each channel signal reconstruct compares; Selected to comprise relatively independently passage of maximum passage of information and information,, can from comprise the maximum passage of information, restore the information of other passage according to the phase space reconfiguration theory.Simultaneously, the present invention proposes the unified algorithm of phase space reconfiguration, when guaranteeing computational accuracy, improved the speed of reconstruct greatly, for the real-time analysis of eeg signal provides assurance; Mathematical model of the present invention is simple, explicit physical meaning, and it is accurate to be easy to realization and to calculate effect.

Below in conjunction with accompanying drawing and embodiment the present invention is further specified.

Description of drawings

The brain wave observation acupuncture point figure of Fig. 1 initial selected;

Fig. 2 is a phase space reconfiguration brain wave analysis operation flow chart of the present invention;

The specific embodiment

The present invention will find out quantitative foundation, gathers acupoint information at first in a large number, has proposed a kind of standard then and has weighed the size of each acupoint information amount, and finally selects one-channel signal and carry out processed steps.

Present embodiment is according to the operating process shown in the accompanying drawing 2, and a simple embodiment of gathering based on the human brain electric wave signal shown in the accompanying drawing 1.

At first can the data that obtain be transferred in the PC through the usb data line, in PC, the transmission type of data be changed, according to acquisition channel storage { x (t through the signal at initially selected 32 acupuncture points of eeg signal acquisition equipment collection _i).

Compare the embedding dimension m value of each passage then, obtain maximum embedding dimension value m _MaxIf, m _Max＞32, increase acquisition target, promptly increase and gather the acupuncture point number, make it equal m _MaxIf, m _Max≤32, m then _MaxThe contained quantity of information of corresponding passage is maximum, and to choose this road be final preferred passage, keeps the passage of m=1 simultaneously; m _MaxBe to comprise the maximum passage of information, m=1 is the relatively independent passage of E.E.G information;

At last, then choose and have the maximum dimension m that embeds _MaxPassage be object of study, keep to embed the passage of dimension m=1 simultaneously; Because m _MaxBe to comprise the maximum passage of information, m=1 is the relatively independent passage of E.E.G information

Proper collection acupuncture point during the final human brain signals collecting that obtains of the present invention; Simplify the complexity of signals collecting through the phase space reconfiguration theory, strengthened the effect of eeg signal real-time analysis, can reconstruct whole brain signal output accurately; The mathematical model of system reconfiguration is simple; Explicit physical meaning is easy to realize that effect is accurate.

Claims (1)

1. the acupuncture point system of selection that brain wave is gathered is characterized in that comprising the steps:

At first, be acquisition target with 32 acupuncture points of brain, make up the acupuncture point passage respectively, through setting-up time length, obtain the eeg signal of 32 acupuncture point passages; The data of 32 acupuncture point passages are stored respectively, obtain the array of 32 equal lengths;

Then, utilize the phase space reconfiguration theory respectively each array to be studied as an independent systematic component, adopt the FNN method to confirm to embed dimension m, adopt the numerical value of first extreme point of AD method as delay time T;

Secondly, each array is analyzed one by one, obtained m ₁... m _i..m ₃₂, m wherein _iThe embedding dimension of representing the data of i degree passage; Through to embedding the screening of relatively studying passage of dimension: if m _Max＞32, m _MaxRepresent that the maximum in all passages embeds dimension, increase and gather the acupuncture point number, make it equal m _MaxIf, m _Max≤32, then choose and have m _MaxPassage be object of study, keep m simultaneously _i=1 passage; Comprise m _MaxPassage be to comprise the maximum passage of information, m _i=1 passage is the relatively independent passage of E.E.G information;

At last, obtain brain wave reconstruction model information: with m _MaxFor embedding dimension, be time delay with its corresponding τ, the reconfiguration system model:

X(0)＝[x(0)，x(τ)，x(2τ)，L，x((m-1)τ)]

X(1)＝[x(1)，x(τ+1)，x(2τ+1)，L，x((m-1)τ+1)]

................................

X(n)＝[x(n)，x(n+τ)，x(n+2τ)，L，x(n+(m-1)τ)]

...............................................................

X(m-1)＝[x(m-1)，x(m-1+τ)，x(m-1+2τ)，L，x((m-1)(τ+1))]

Subsequent analysis is only gathered m _MaxWith the m=1 channel data.

Images