CN105193409A - Method and system for evaluating electroencephalogram inhibition level - Google Patents

Abstract

The application discloses a method for evaluating the electroencephalogram inhibition level. The method comprises a process of calculating the index of an electroencephalogram inhibition level, wherein the process comprises the steps of dividing an electroencephalogram signal into segments, removing noise in the electroencephalogram signal, determining a stroke signal segment and an inhibition signal segment in the rest segments, and calculating the index of the inhibition level according to the proportion of the number of inhibition signal segments in the number of segments without noise segments. The application further discloses a system for evaluating the electroencephalogram inhibition level. According to the method and the system, the electroencephalogram signal is divided into segments, and the index of the inhibition level is calculated according to the proportion of the number of inhibition signal segments in the number of the segments without noise segments, so that noise can be effectively removed since the noise segments are removed, the accuracy for calculating the index of the electroencephalogram inhibition level is improved, and the application value of the parameter is promoted.

Description

A kind of brain electricity suppresses level evaluation method and system

Technical field

The application relates to medical apparatus and instruments, particularly relates to a kind of brain electricity and suppresses level evaluation method and system.

Background technology

EEG (Electroencephalograph, electroencephalogram) outburst suppresses signal to be a kind of EEG signals feature that (such as under deep anaesthesia) just there will be under special state, just gradually studied in the last few years.EEG outburst suppresses the brain electricity of the present angle value by a narrow margin of signal waveform mark sheet to suppress the brain electric detonation signaling states of signal condition and high frequency high-amplitude alternately to occur, and this alternate is not periodically.Be illustrated in figure 1 one section of outburst and suppress signal.E is wherein noise, and F is detonator signal, and G is for suppressing signal, and when breaking out inhibitory state and occurring, the metabolic rate of brain will minimize.There is outburst inhibitory state if long-time, illustrates that patient is in the sign state of danger, easily cause the impaired or brain death of brain.By suppressing the analysis of level to EEG signal, effectively can evaluate the excitement degree of brain, in evaluation patient level of consciousness, depth of anesthesia, there is very important using value.Current EEG signal suppresses assessment of levels mainly to utilize BSR (BurstSuppressionRatio, outburst rejection ratio index), it is mainly through evaluating in certain hour section, the signal segment that the absolute value of EEG signal amplitude is no more than 5 microvolts (uV) accounts for the ratio of total time section, but this simply by amplitude judge EEG signal always level method easily by noise erroneous judgement be decided to be detonator signal, cause accuracy and reliability bad.

Current evaluation brain electricity suppression level mainly utilizes BSR.The multi-parameter monitor with BIS (BispectralIndex, depth of anesthesia) monitoring function gives following definition to BSR parameter: the ratio of total time shared by the time that in 60S, EEG signals considered to be in inhibitory state in the past.Rampil etc. give also the computational methods of outburst rejection ratio.Be defined in and be greater than signal duration on the sub-time period of 0.5S, the absolute value of EEG signals amplitude is not more than 5uV and is namely judged to suppress signal, thus can calculate outburst rejection ratio parameter.But the anti-interference wretched insufficiency of the method, if EEG signal is subject to various interference (such as body is dynamic, eye is electric, myoelectricity, electric knife, defibrillation etc.), noise can be easy to be judged to be detonator signal, causes the not accurate enough property of above-mentioned parameter result of calculation.

Adopt fixed threshold method in prior art, if the baseline drift of brain electricity (EEG) signal, can cause and cannot judge that outburst suppresses signal, can not intelligent decision.And the threshold value real-time update of the application, more accurately.

Step 1064: if the standard deviation corresponding to segmentation EEG signals is greater than first threshold, then judge that this section of EEG signals is noise.

Step 108: distinguish the suppression signal in remaining segmentation EEG signals and detonator signal.

In one embodiment, distinguish the suppression signal in remaining segmentation EEG signals and detonator signal, specifically comprise the following steps:

Step 1082: according to the intermediate value of segmentation EEG signals standard deviation and the maximum value calculation Second Threshold Threshold2 of segmentation EEG signals standard deviation.

(4) Second Threshold Threshold2 specifically calculates by formula:

Thresold2 (j)=(1-δ) * Thresold2 (j-1)+α * med formula (4)

Wherein, j=1,2 ..., N;

Wherein Thresold2 (j-1) is the decision threshold of the data suppression signal of a upper mS, and Thresold2 (j) is the decision threshold of these mS data.Med and max is this mS data sectional, removes intermediate value and the maximum of the standard deviation that noise section is left afterwards.δ is weights, 0 < δ < 1.δ equals the ratio of med and max.Utilize the threshold value of these weights and the preceding paragraph mS data sectional standard deviation, upgrade the threshold value obtaining this segment data standard deviation.

Step 1084: if the standard deviation corresponding to segmentation EEG signals is less than Second Threshold, then judge that this section of EEG signals is as suppressing signal, otherwise this section of EEG signals is detonator signal.

Step 110: by suppressing the hop count of signal to account for the ratio suppressing signal and the total hop count of detonator signal, calculating and suppressing level index.

formula (5)

Wherein, suppress time hop count actual for suppressing the hop count of signal, outburst suppresses the actual hop count for suppressing the hop count of signal to add detonator signal of total time section section.

Get the EEG signals that a section adds the interference such as electric knife, calculate EEG signals every segment standard difference cloth as shown in Figure 4, wherein A is noise (i.e. noise segment) corresponding standard deviation, B is detonator signal (i.e. detonator signal section) corresponding standard deviation, C, for suppressing signal (namely suppressing signal segment) corresponding standard deviation, obviously can find out noise, detonator signal, suppression signal standards difference is furnished with obvious difference.

The application is by after Signal Pretreatment, carry out noise artefact decision-making function effectively, effectively can remove noise, and realize the self-adaptative adjustment of threshold value, add the accuracy that EEG signals suppresses the calculating of level index, promote the using value of this parameter.

Summary of the invention

The application provides a kind of brain electricity to suppress level evaluation method and system.

According to the first aspect of the application, the application provides a kind of brain electricity to suppress level evaluation method and system, comprises the process calculating brain electricity and suppress level index, and described calculating brain electricity suppresses the process of level index to comprise:

By EEG signals segmentation, removal noise segment wherein, judges the detonator signal in residue hop count and suppression signal, and the ratio of the hop count after except noise segment that then accounts for according to suppression signal hop count, calculates suppression level index.

Said method, described calculating brain electricity suppresses the process of level index specifically to comprise:

Gather the EEG signals of nearest scheduled duration, described EEG signals is divided into multistage;

Calculate the intermediate value of segmentation EEG signals standard deviation and the maximum of described segmentation EEG signals standard deviation;

Judge whether described segmentation EEG signals is noise, if then remove according to described intermediate value and described maximum;

Distinguish the suppression signal in remaining described segmentation EEG signals and detonator signal;

Accounted for the ratio suppressing signal and the total hop count of detonator signal by the hop count of described suppression signal, calculate and suppress level index.

According to intermediate value and described maximum, said method, describedly judges whether described segmentation EEG signals is noise, specifically comprises:

According to described intermediate value and described maximum value calculation first threshold Threshold1;

$Threshold1\left(j\right)={\mathrm{\&lambda;}}_1*\frac{(Med+Max)}{2}+(1-{\mathrm{\&lambda;}}_1)*Threshold1(j-1)$

Wherein, ${\mathrm{\&lambda;}}_1=\left\{\begin{array}{cc}\frac{Med}{Threshold1\left(j-1\right)}& Med\&le;Threshold1\left(j-1\right)\\ \frac{Threshold1\left(j-1\right)}{Med}& Med>Threshold1\left(j-1\right)\end{array}\right.$

Wherein, j=1,2 ..., N; Med is the intermediate value of described segmentation EEG signals standard deviation, and Max is the maximum of described segmentation EEG signals standard deviation; N is the number of times that data segment upgrades.

If the standard deviation corresponding to described segmentation EEG signals is greater than first threshold, be then judged as noise.

Said method, the suppression signal in the remaining described segmentation EEG signals of described differentiation and detonator signal, specifically comprise:

According to described intermediate value and described maximum value calculation Second Threshold Threshold2;

Thresold2(j)＝(1-δ)*Thresold2(j-1)+α*med

Wherein, j=1,2 ..., N;

If the standard deviation corresponding to described segmentation EEG signals is less than Second Threshold, is then judged as suppressing signal, otherwise is detonator signal.

Said method, described intermediate value, obtains especially by following steps:

Calculate the standard deviation Std of described segmentation EEG signals;

wherein x _kn () is segmentation EEG signals, for the average of segmentation EEG signals, L is the number of data points of segmentation EEG signals; K=1,2 ..., M is the sequence of data segment, and M is the hop count of data segment;

Described intermediate value is calculated according to described standard deviation Std;

Med=median{Std (k) } _{k=1,2 ..., M}, wherein Med is described intermediate value.

According to the second aspect of the application, the application provides a kind of brain electricity to suppress proficiency assessment system, described system, for by EEG signals segmentation, remove noise segment wherein, judge the detonator signal in residue hop count and suppression signal, the ratio of the hop count after except noise segment that then accounts for according to suppression signal hop count, calculates suppression level index.

Described system comprises:

Acquisition module, for gathering the EEG signals of nearest scheduled duration, is divided into multistage by described EEG signals;

Computing module, for the maximum of the intermediate value and described segmentation EEG signals standard deviation that calculate segmentation EEG signals standard deviation;

According to described intermediate value and described maximum, judge module, for judging whether described segmentation EEG signals is noise, if then remove; Distinguish the suppression signal in remaining described segmentation EEG signals and detonator signal;

Processing module, for being accounted for the ratio suppressing signal and the total hop count of detonator signal by the hop count of described suppression signal, is calculated and suppresses level index.

Described judge module also for:

According to described intermediate value and described maximum value calculation first threshold Threshold1;

$Threshold1\left(j\right)={\mathrm{\&lambda;}}_1*\frac{(Med+Max)}{2}+(1-{\mathrm{\&lambda;}}_1)*Threshold1(j-1)$

Wherein, ${\mathrm{\&lambda;}}_1=\left\{\begin{array}{cc}\frac{Med}{Threshold1\left(j-1\right)}& Med\&le;Threshold1\left(j-1\right)\\ \frac{Threshold1\left(j-1\right)}{Med}& Med>Threshold1\left(j-1\right)\end{array}\right.$

Wherein, j=1,2 ..., N; Med is the intermediate value of described segmentation EEG signals standard deviation, and Max is the maximum of described segmentation EEG signals standard deviation; N is the number of times that data segment upgrades.

If the standard deviation corresponding to described segmentation EEG signals is greater than first threshold, be then judged as noise.

Described judge module also for:

According to described intermediate value and described maximum value calculation Second Threshold Threshold2;

Thresold2(j)＝(1-δ)*Thresold2(j-1)+α*med

Wherein, j=1,2 ..., N;

If the standard deviation corresponding to described segmentation EEG signals is less than Second Threshold, is then judged as suppressing signal, otherwise is detonator signal.

Described computing module also for:

Calculate the standard deviation Std of described segmentation EEG signals;

$Std\left(k\right)=\sqrt{\frac{{\mathrm{\&Sigma;}}_{n=1}^L{\left(x_k\right(n)-{\stackrel{\&OverBar;}{x}}_k)}^2}{L}},$

Wherein x _kn () is segmentation EEG signals, for the average of segmentation EEG signals, L is the number of data points of segmentation EEG signals; K=1,2 ..., M is the sequence of data segment, and M is the hop count of data segment;

Med=median{Std (k) } _{k=1,2 ..., M}, wherein Med is described intermediate value.

Owing to have employed above technical scheme, the beneficial effect that the application is possessed is:

In the detailed description of the invention of the application, by EEG signals segmentation, then according to suppressing signal hop count to account for the ratio of the hop count after except noise segment, calculate suppression level index, owing to removing noise segment wherein, effectively can remove noise, add the accuracy that EEG signals suppresses the calculating of level index, promote the using value of this parameter.

Accompanying drawing explanation

Fig. 1 is EEG signals schematic diagram;

Fig. 2 is the method flow chart in one embodiment of the application;

Fig. 3 is the method for the application, the renewal when EEG process, fragmentation procedure schematic diagram;

Fig. 4 is the EEG signals block signal standard difference Butut after adding noise;

Fig. 5 is the system high-level schematic functional block diagram in one embodiment of the application.

Detailed description of the invention

By reference to the accompanying drawings the application is described in further detail below by detailed description of the invention.

Embodiment one:

The application provides a kind of brain electricity to suppress level evaluation method, comprise the process calculating brain electricity and suppress level index, calculating brain electricity suppresses the process of level index to comprise: by EEG signals segmentation, remove noise segment wherein, judge the detonator signal in residue hop count and suppression signal, then according to suppressing signal hop count to account for the ratio of the hop count after except noise segment, suppression level index is calculated.

As shown in Figure 2, the calculating brain electricity of the application suppresses the process of level index specifically to comprise the following steps:

Step 102: the EEG signals gathering nearest scheduled duration, is divided into multistage by EEG signals.EEG signals after segmentation, i.e. segmentation EEG signals, is actually one section of EEG signals section.

EEG signals after traditional classical filter process effectively can get rid of the noise beyond frequency band range, but for larger interference, such as body moves, eye electricity, myoelectricity, electric knife etc., to rely on conventional filtering to remove, the application proposes a kind of method removing brain electricity (EEG) signal fluctuation noise, by brain electricity (EEG) signal is divided into some sub-time periods, the time span of concrete the hop count that divides and every section can be arranged based on experience value, judge whether each section be noise, if it is determined that be noise, then do not participate in brain electricity (EEG) parameter to calculate.

Step 104: calculate the intermediate value of segmentation EEG signals standard deviation and the maximum of segmentation EEG signals standard deviation.

In one embodiment, the intermediate value of segmentation EEG signals standard deviation, obtains especially by following steps:

Step 1042: the standard deviation Std calculating segmentation EEG signals.

Specifically (1) calculate according to formula:

$Std\left(k\right)=\sqrt{\frac{{\mathrm{\&Sigma;}}_{n=1}^L{\left(x_k\right(n)-{\stackrel{\&OverBar;}{x}}_k)}^2}{L}}$ Formula (1)

Wherein x _kn () is segmentation EEG signals, for the average of segmentation EEG signals, L is the number of data points of segmentation EEG signals; K=1,2 ..., M is the sequence of data segment, and M is the hop count of data segment.

When Fig. 3 is EEG process, renewal process and fragmentation procedure schematic diagram, get the data segment of 60S (S is unit of time: second, lower same), this data segment be further divided into the subsegment of 0.5S, so one has 120 subsegments, i.e. M=120.The method of this data sectional is empirical value, also can divide by additive method, and the data hop count obtained like this is not necessarily 120.The data subsegment of every section of 0.5S, when sample rate is 256Hz, L=256*0.5=128, namely the data point of each subsegment has 128.Here L-value obtains according to the sample rate of data and the length of data sectional, different EEG sample frequency and data segmentation method, the result obtained and L value also different.

Step 1044: the intermediate value calculating segmentation EEG signals standard deviation according to standard deviation Std.

Intermediate value Med represents, specifically (2) calculates by formula:

Med=median{Std (k) } _{k=1,2 ..., M}formula (2)

The wherein M hop count that divide to by EEG signals.

Step 106: judge whether segmentation EEG signals is noise, if then remove according to the intermediate value of segmentation EEG signals standard deviation and the maximum of segmentation EEG signals standard deviation.

In one embodiment, judge whether segmentation EEG signals is noise, specifically comprises the following steps according to the intermediate value of segmentation EEG signals and the maximum of segmentation EEG signals:

Step 1062: according to the intermediate value of segmentation EEG signals standard deviation and the maximum value calculation first threshold Threshold1 of segmentation EEG signals standard deviation.

(3) first threshold Threshold1 specifically calculates by formula:

$Threshold1\left(j\right)={\mathrm{\&lambda;}}_1*\frac{(Med+Max)}{2}+(1-{\mathrm{\&lambda;}}_1)*Threshold1(j-1)$ Formula (3)

Wherein, ${\mathrm{\&lambda;}}_1=\left\{\begin{array}{cc}\frac{Med}{Threshold1\left(j-1\right)}& Med\&le;Threshold1\left(j-1\right)\\ \frac{Threshold1\left(j-1\right)}{Med}& Med>Threshold1\left(j-1\right)\end{array}\right.$

Wherein, j=1,2 ..., N; Med is the intermediate value of described segmentation EEG signals standard deviation, and Max is the maximum of described segmentation EEG signals standard deviation; N is the number of times that data segment upgrades.

Wherein Thresold1 (j-1) is the data judging threshold value of a upper mS; λ ₁weights; Thresold1 (j) is the decision threshold of these mS data.Max and Med is the maximum of the standard deviation corresponding to data segment after this (60S) data sectional and the intermediate value of standard deviation.

Embodiment two:

The brain electricity of the application suppresses proficiency assessment system, its a kind of embodiment, described system, for by EEG signals segmentation, remove noise segment wherein, judge the detonator signal in residue hop count and suppression signal, the ratio of the hop count after except noise segment that then accounts for according to suppression signal hop count, calculates suppression level index.

As shown in Figure 5, the brain electricity of the application suppresses proficiency assessment system to comprise acquisition module, computing module, judge module and processing module.Acquisition module, for gathering the EEG signals of nearest scheduled duration, is divided into multistage by EEG signals; Computing module, for the maximum of the intermediate value and described segmentation EEG signals standard deviation that calculate segmentation EEG signals standard deviation; According to the intermediate value of segmentation EEG signals standard deviation and the maximum of segmentation EEG signals standard deviation, judge module, for judging whether segmentation EEG signals is noise, if then remove; Distinguish the suppression signal in remaining segmentation EEG signals and detonator signal; Processing module, for by suppressing the hop count of signal to account for the ratio suppressing signal and the total hop count of detonator signal, calculating and suppressing level index.

In one embodiment, judge module is also for according to the intermediate value of segmentation EEG signals standard deviation and the maximum value calculation first threshold Threshold1 of segmentation EEG signals standard deviation.

$Threshold1\left(j\right)={\mathrm{\&lambda;}}_1*\frac{(Med+Max)}{2}+(1-{\mathrm{\&lambda;}}_1)*Threshold1(j-1)$

Wherein, ${\mathrm{\&lambda;}}_1=\left\{\begin{array}{cc}\frac{Med}{Threshold1\left(j-1\right)}& Med\&le;Threshold1\left(j-1\right)\\ \frac{Threshold1\left(j-1\right)}{Med}& Med>Threshold1\left(j-1\right)\end{array}\right.$

Wherein, j=1,2 ..., N; Med is the intermediate value of described segmentation EEG signals standard deviation, and Max is the maximum of described segmentation EEG signals standard deviation; N is the number of times that data segment upgrades.

Judge module is also greater than first threshold for the standard deviation corresponding to segmentation EEG signals, then this segmentation EEG signals is judged as noise.

The brain electricity of the application suppresses proficiency assessment system, and judge module is also for according to the intermediate value of segmentation EEG signals and the maximum value calculation Second Threshold Threshold2 of segmentation EEG signals.

Thresold2(j)＝(1-δ)*Thresold2(j-1)+α*med

Wherein, j=1,2 ..., N;

Judge module is also less than Second Threshold for the standard deviation corresponding to segmentation EEG signals, then judge that this segmentation EEG signals is as suppressing signal, otherwise be detonator signal.

In one embodiment, computing module is also for calculating the standard deviation Std of segmentation EEG signals.

Wherein, x _kn () is segmentation EEG signals, for the average of segmentation EEG signals, N is the number of data points of segmentation EEG signals.

Computing module also can be used for the intermediate value calculating segmentation EEG signals standard deviation according to standard deviation Std.

Med=median{Std (k) } _{k=1,2 ..., M}, the wherein M hop count that divide to by EEG signals.

Above content is the further description done the application in conjunction with concrete embodiment, can not assert that the concrete enforcement of the application is confined to these explanations.For the application person of an ordinary skill in the technical field, under the prerequisite not departing from the application's design, some simple deduction or replace can also be made.

Claims (10)

1. brain electricity suppresses a level evaluation method, it is characterized in that, comprises the process calculating brain electricity and suppress level index, and described calculating brain electricity suppresses the process of level index to comprise:

By EEG signals segmentation, removal noise wherein, judges the detonator signal in residue hop count and suppression signal, and the ratio of the hop count after except noise that then accounts for according to suppression signal hop count, calculates suppression level index.

2. brain electricity as claimed in claim 1 suppresses level evaluation method, it is characterized in that, described calculating brain electricity suppresses the process of level index specifically to comprise:

Gather the EEG signals of nearest scheduled duration, described EEG signals is divided into multistage;

Calculate the intermediate value of segmentation EEG signals standard deviation and the maximum of described segmentation EEG signals standard deviation;

Judge whether segmentation EEG signals is noise, if then remove according to the intermediate value of described standard deviation and the maximum of described standard deviation;

Distinguish the suppression signal in remaining described segmentation EEG signals and detonator signal;

Accounted for the ratio suppressing signal and the total hop count of detonator signal by the hop count of described suppression signal, calculate and suppress level index.

3. brain electricity as claimed in claim 2 suppresses level evaluation method, it is characterized in that,

The described intermediate value according to described EEG signals standard deviation and maximum judge whether described segmentation EEG signals is noise, specifically comprises:

According to described standard deviation intermediate value and described standard deviation maximum value calculation first threshold Threshold1;

$Threshold1\left(j\right)={\mathrm{\&lambda;}}_1*\frac{(Med+Max)}{2}+(1-{\mathrm{\&lambda;}}_1)*Threshold1(j-1)$

Wherein, ${\mathrm{\&lambda;}}_1=\left\{\begin{array}{cc}\frac{Med}{Threshold1\left(j-1\right)}& Med\&le;Threshold1\left(j-1\right)\\ \frac{Threshold1\left(j-1\right)}{Med}& Med>Threshold1\left(j-1\right)\end{array}\right.,$ J=1,2 ..., N; Med is the intermediate value of described segmentation EEG signals standard deviation, and Max is the maximum of described segmentation EEG signals standard deviation; N is the number of times of Data Update.

If the standard deviation corresponding to described segmentation EEG signals is greater than first threshold, be then judged as noise.

4. brain electricity as claimed in claim 3 suppresses level evaluation method, and it is characterized in that, the suppression signal in the remaining described segmentation EEG signals of described differentiation and detonator signal, specifically comprise:

According to described intermediate value and described maximum value calculation Second Threshold Threshold2;

Thresold2(j)＝(1-δ)*Thresold2(j-1)+α*med

Wherein, j=1,2 ..., N;

If the standard deviation corresponding to described segmentation EEG signals is less than Second Threshold, is then judged as suppressing signal, otherwise is detonator signal.

5. the brain electricity as described in claim 3 or 4 suppresses level evaluation method, it is characterized in that described intermediate value obtains especially by following steps:

Calculate the standard deviation Std of described segmentation EEG signals;

wherein x _kn () is segmentation EEG signals, for the average of segmentation EEG signals, L is the number of data points of segmentation EEG signals; K=1,2 ..., M is the sequence of data segment, and M is the hop count of data segment;

Described intermediate value is calculated according to described standard deviation Std;

Med=median{Std (k) } _{k=1,2 ..., M}, wherein Med is described intermediate value.

6. brain electricity suppresses a proficiency assessment system, it is characterized in that,

Described system, for by EEG signals segmentation, removes noise segment wherein, judges detonator signal in residue hop count and suppresses signal, then according to suppressing signal hop count to account for the ratio of the hop count after except noise segment, calculates suppression level index.

7. brain electricity as claimed in claim 6 suppresses proficiency assessment system, and it is characterized in that, described system comprises:

Acquisition module, for gathering the EEG signals of nearest scheduled duration, is divided into multistage by described EEG signals;

Computing module, for the maximum of the intermediate value and described segmentation EEG signals standard deviation that calculate segmentation EEG signals standard deviation;

According to described intermediate value and described maximum, judge module, for judging whether described segmentation EEG signals is noise, if then remove; Distinguish the suppression signal in remaining described segmentation EEG signals and detonator signal;

Processing module, for being accounted for the ratio suppressing signal and the total hop count of detonator signal by the hop count of described suppression signal, is calculated and suppresses level index.

8. brain electricity as claimed in claim 7 suppresses proficiency assessment system, it is characterized in that,

Described judge module also for:

According to described intermediate value and described maximum value calculation first threshold Threshold1;

$Threshold1\left(j\right)={\mathrm{\&lambda;}}_1*\frac{(Med+Max)}{2}+(1-{\mathrm{\&lambda;}}_1)*Threshold1(j-1)$

Wherein, ${\mathrm{\&lambda;}}_1=\left\{\begin{array}{cc}\frac{Med}{Threshold1\left(j-1\right)}& Med\&le;Threshold1\left(j-1\right)\\ \frac{Threshold1\left(j-1\right)}{Med}& Med>Threshold1\left(j-1\right)\end{array}\right.$

Wherein, j=1,2 ..., N; Med is the intermediate value of described segmentation EEG signals standard deviation, and Max is the maximum of described segmentation EEG signals standard deviation;

If the standard deviation corresponding to described segmentation EEG signals is greater than first threshold, be then judged as noise.

9. brain electricity as claimed in claim 8 suppresses proficiency assessment system, it is characterized in that, described judge module also for:

According to described intermediate value and described maximum value calculation Second Threshold Threshold2;

Thresold2(j)＝(1-δ)*Thresold2(j-1)+α*med

Wherein, j=1,2 ..., N;

If the standard deviation corresponding to described segmentation EEG signals is less than Second Threshold, is then judged as suppressing signal, otherwise is detonator signal.

10. as claimed in claim 8 or 9 brain electricity suppresses proficiency assessment system, it is characterized in that, described computing module also for:

Calculate the standard deviation Std of described segmentation EEG signals;

wherein x _kn () is segmentation EEG signals, for the average of segmentation EEG signals, L is the number of data points of segmentation EEG signals; K=1,2 ..., M is the sequence of data segment, and M is the hop count of data segment;

Described intermediate value is calculated according to described standard deviation Std;

Med=median{Std (k) } _{k=1,2 ..., M}, wherein Med is described intermediate value.