JP2006255134A - Brain wave measurement/display method and device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable a measurement, a display and a control of a state corresponding to every sedation and consciousness level from a shallow depth to a deep depth. <P>SOLUTION: This brain wave measurement/display device is provided with: a brain wave information processing means 2 obtaining brain wave information 1 indicating a correlation between an index formed of various types of brain wave information entropy created from the measured brain waves and information of brain wave complexity, the level of drug in the blood, and anesthetic depth, and second brain wave state information from the measured brain waves such as brain wave information entropy, a power spectrum and brain wave complexity information; a linear combination means 3 giving an index (ci) of the sedation/consciousness level from the product (xi×ki) of the disorder (xi) of the brain wave and a weighting factor (ki) corresponding to anesthetic (i) (i=1, 2,...N) and giving an index C in a specified sedation/consciousness level by Σci; a determination means 4 determining the anesthetic depth as a function of the administered drug concentration based on the brain wave information in real time, the linear combination means and the second brain wave state information in real time; a control means 5 optimizing by controlling the drug administration condition from the determination result; and a display means 6 displaying the determination and the state of the control. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an electroencephalogram information measurement process and a method and apparatus for displaying the process result.

 An electroencephalogram contains a great deal of information at random in an arousal state, and at first glance it is measured as random noise, and it is considered impossible to extract information from the electroencephalogram in this state. However, since only very limited information is included in a sedated state due to anesthesia or the like, the electroencephalogram is adopted as an index representing the state of the brain. Therefore, an electroencephalogram is often used as a method for determining the depth of anesthesia. In these quantitative methods, the characteristics of the electroencephalogram are extracted and quantified by various analysis methods. The power spectrum, the intermediate frequency (MF), the spectrum edge frequency 95 (SF95), the delta power (DP), etc., obtained by Fourier transform of the electroencephalogram are well known as classical analysis methods.

These electroencephalogram analysis techniques are described in Non-Patent Document 1. In the conventional electroencephalogram waveform analysis, brain potentials are recorded on paper with a strip chart recorder and then visually observed to extract the characteristics. Since the standard recording speed is 25 mm per second, in the electroencephalogram analysis for about 1 hour, it was necessary to analyze the recording of 90 m in detail. Frequency analysis is the first method that has shifted from visual EEG measurement to analysis EEG measurement. In order to extract the regularity included in the electroencephalogram, the raw waveform is separated into a harmonic frequency component (Fourier transform) of a trigonometric function (frequency spectrum), and the square of the amplitude (power spectrum) of the frequency component is obtained. These spectra are called delta (0.5-3.4Hz), theta (3.5-7.4Hz), sub-alpha (7.5-9.4Hz), alpha (9.5-12.4Hz), beta (12.5-30Hz) waves from the low frequencies. Since the amplitude tends to increase in the delta region as the sedation level progresses, the absolute value of the power in the delta region is called absolute delta power (delta power), and the ratio to the total frequency region power is called relative delta power. On the other hand, the median frequency is a frequency indicating an intermediate value of the amplitude distribution over all these frequency ranges. A spectral edge frequency 95 (SF95) is a frequency when 95% of the total power is counted from the low frequency range.
IJ Rampil, "A primer for EEG signal processing in anesthesia, Anethesiology, (1998), 89, 4, pp.980-1002

  These indices tend to increase delta power and decrease SF95 and intermediate frequencies as the sedation level progresses. However, there is a large variance among the targeted patient groups and the correlation with the depth of anesthesia is low, and its clinical use tends to decrease. The classical analysis method assumed EEG data to be continuous regular data with internal regularity (deterministic law such as repetition) based on causality, and tried to quantify the period. Therefore, more advanced techniques such as bispectral analysis (BIS), entropy such as approximate entropy (APEN), Shannon's information entropy (SHEN), spectral entropy (SPEEN), and complexity (COMPLEX) measurement methods are being introduced in recent years. is there. Entropy and complexity are attracting attention in algorithms published in many documents such as Anesthesia as a new analysis method. The electroencephalogram includes regularities of various ranks. When the regularity is high, there are few types of waveform patterns constituting the electroencephalogram, and when the regularity is low, the number of waveform patterns constituting the electroencephalogram is large.

  The waveform pattern that composes an electroencephalogram is like a state variable (entropy) in thermodynamics and is called entropy. In the case of an electroencephalogram, if the regularity is high, the entropy is low, and conversely, if the regularity is low, the entropy is high. To do. The electroencephalogram information entropy is described in the aforementioned Non-Patent Document 1. Entropy is also called irregularity because it shows a change opposite to regularity. There is approximate entropy as a method for calculating irregularities. On the other hand, what extracted the amplitude of the electroencephalogram, calculated | required the frequency for every amplitude (frequency distribution of amplitude), and calculated | required those Shannon (Shannon) information entropy is called amplitude entropy (amplitude entropy). Further, spectral entropy is obtained by frequency-analyzing the electroencephalogram, obtaining the frequency for each frequency (frequency frequency distribution), and similarly obtaining Shannon's information entropy. Depending on the target frequency range, there are state entropy (SE) and response entropy (RE). The frequency ranges are 0.8-32Hz and 0.8-47Hz, respectively. The latter includes electroencephalogram and electromyogram. It is said that the electromyogram changes in the shallow sedation level, especially in the situation just before awakening, indicate imminent awakening.

  The regularity of the electroencephalogram can be regarded as the complexity of the electroencephalogram. In this case, when the regularity is high, the complexity is considered low (simple), and conversely, when the regularity is low, the complexity is considered high. When estimating the future value of the electroencephalogram from the current value, regularity is an important clue, and when there is sufficient regularity, these relations can be understood because they can be estimated easily (= simply). In such an autoregressive model (deterministic process), complexity is a measure for its determination. There are neural complexity, KL-complexity, and Lempel-Ziv complexity as quantitative measures of complexity. Especially in Lempel-Ziv, the calculation method is simple and convenient for clinical use.

  Such an electroencephalogram analysis by a public algorithm has just started in recent years, and it is concluded that the correlation result with the depth of anesthesia is not constant. However, since the algorithm is open to the public, it can be verified by many facilities and researchers.

  On the other hand, algorithms (BIS, ARX, PSI, Narcotrend) based on new undisclosed analysis methods are also incorporated in the electroencephalograph. For example, the BIS monitor of Aspect Medical Systems, USA is well known. In this frequency analysis, a power spectrum that is a square value of the amplitude is used, so the phase of the spectrum is lost and only amplitude information is obtained. In contrast, bispectral analysis uses phase information. The electroencephalogram has different frequency components in each part by frequency analysis. Let's now assume that the phase difference between the frequency component a1 at location A and the frequency component b1 at location B is called a1 and b1 coupling. From the combination of (a1..aN) × (b1..bN) from all frequency components a1..aN at location A and all frequency components b1..bN at location B, a total of N2 A set of phase difference coupling (bispectral set) can be formed. BIS (bispectral index) represents this set as a single index from 0 to 100. The process for this indicator is not disclosed. The BIS monitor has an internal database (statistical template) obtained from a large number of clinical data, and the measurement data is referred to this template and then converted to BIS. Other private algorithms include ARX, PSI, and Narcotrend. About the outline of the analysis method by these private algorithms, the prospect of practical use is unclear at present.

  Among the above-mentioned non-public algorithms, the BIS monitor that has been commercially available for the longest time as an anesthesia depth monitor has a high correlation between PD and a moderate anesthesia level, and a low correlation with a shallow anesthesia level. In addition, at medium to deep levels of anesthesia, saturation of BIS is observed due to the appearance of burst suppression waves. Moreover, when using nitrous oxide gas during surgery, there is a high probability that it will be mistaken as a shallow or deep (divided by researchers) level than the actual depth level. Another problem is that only certain sedatives have a selective effect, such as sedatives that exhibit high correlation while others have only low correlation.

  On the other hand, in recent studies, it is often assumed that the electroencephalogram is a signal derived from a stochastic irregular process, and is analyzed with the intention of discussing the degree of regularity in these irregular processes. The mathematical theories of these conventional and conventional analysis methods are established and clear, but what are the characteristics of these calculation results in various data environments, which are superior in which environments, etc. Is not clear. It is concluded that these will be finally clarified in actual clinical use, but it is estimated that it will take time to derive the results in clinical environments with different conditions and factors such as various anesthesia, surgical technique and pathology. .

  In the prior art that measures, displays, and manages the depth of anesthesia using the index obtained from the above-described electroencephalogram information, there are many limitations between the type of anesthetic and the feasible sedation level. Therefore, it is impossible to measure, display, and manage the depth of anesthesia in response to sedation and consciousness levels of all depths from shallow to deep, which hinders control of sedation and consciousness levels. Therefore, it is an object of the present invention to remove such restrictions and enable measurement, display, and management of states corresponding to all sedation and consciousness levels from shallow to deep.

  In order to solve the above problems, in the present invention, the sedation and consciousness level from shallow to deep are defined in a range where the efficacy of a plurality of anesthetics exists in a plurality of ranges, and anesthetic i (i = 1 , 2, …… N), the sedation / consciousness level index ci is given by the product xi · ki of the electroencephalogram randomness xi and the weighting factor ki for each anesthetic, and at the specified sedation / consciousness level. The index C is given by the linear combination Σci of each ci. By converting the logic thus obtained into firmware, an electroencephalogram information measurement process corresponding to an arbitrary sedation / consciousness level from shallow to deep, and a display method and apparatus for the process result are provided.

  By weighting multiple EEG randomness expressed by various entropies generated as EEG state information and the complexity that regulates the regularity of EEG, it is linearly combined, and using this as an index to give sedation and consciousness level, Since the sedation and consciousness level can be controlled to any value from shallow to deep, the relationship between pharmacokinetics and drug efficacy during surgery can be specified easily and quickly over a wide range, and the accuracy of anesthesia control during surgery can be improved. effective. Moreover, since the analysis accuracy of electroencephalogram information is improved by accumulating electroencephalogram analysis data obtained through many clinical controls, it can be used as a monitor for sleep apnea syndrome.

  Next, the present invention will be described in detail. First, before developing the first-order linear theory in the present invention, the relationship of indices to electroencephalogram irregularities will be described by an algorithm. First, a pseudo brain wave signal with unified generation conditions and generation factors is created to evaluate the characteristics of each analysis method. For evaluation, the following two types of test bench models are created and simulated. The complexity of the electroencephalogram uses approximate entropy, Shannon's amplitude entropy, (power) spectral entropy (including state entropy and reaction entropy), Rampelchev complexity, spectral edge 95, intermediate frequency, and the like as indices.

(1) Deterministic autoregressive model (linear model)
Assuming deterministic generation in which the background of brain wave generation is defined by causality, create a calculation model in which the proportion of irregular processes increases during the generation process, and analyze the analysis method by each algorithm. evaluate. When brain wave generation follows this model, the current brain wave can be predicted from past data, and becomes complicated when the prediction formula is higher order. In the simulation, the increase in complexity is controlled by increasing the ratio of the random number sequence.

(2) Probabilistic model (nonlinear dynamic model)
The background of brain wave generation is a nonlinear dynamic phenomenon defined by probability theory, and a logistic map function is used as the brain wave. The configuration parameters were changed and the attractor was bifurcated to the power of 2, 4... 2N, and finally made chaos. We try analysis methods by various algorithms using the function value generated in each convergence state as an electroencephalogram, and evaluate the value of the obtained index. When EEG generation follows this model, the convergence point changes depending on the initial value and the variable value in the equation. The increase in complexity is controlled by increasing the convergence point.

  The simulation method is based on the following method 1 and method 2. In Method 1, the signal that repeats maximum (100) and minimum (0) is a complete repeat signal, and random data consisting of random numbers is increased to 0-100% in increments of 20%. The data string was assumed to be an electroencephalogram. Method 1 corresponds to a deterministic autoregressive model (linear model). The degree of randomness of the electroencephalogram waveform is considered as an index indicating brain arousal. Therefore, the approximate entropy, the Lampelchef complexity, the amplitude entropy, the spectral entropy, the spectral edge 95, and the intermediate frequency are used as indices, and the relationship between the index value and the randomness is plotted as a linear quantity by simulation. FIG. 1 is a plot showing a simulation result by Method 1. As a result, spectral entropy showed the best response.

In Method 2,
Xn + 1 = A * Xn * (1-Xn) (1)
The logistic map function is used. The initial value was 0.25, and the coefficient A was 3.2, 3.48, 3.56, 3.568, 3.57, 3.8 (convergence points: the numbers of attracters were 2, 4, 8, 16, 32, and Chaos, respectively). The number of data generation was 1000.
FIG. 2 shows a simulation result by the method 2. A simulation similar to FIG. 1 was applied to a stochastic model (nonlinear dynamic model) and plotted in FIG. 2 by simulation. As in the case of FIG. 1, approximate entropy, Lampelchef complexity, amplitude entropy, spectral entropy, spectral edge 95, and intermediate frequency were adopted as indices. The index value for the number of branches is obtained by nonlinear dynamics and shown in FIG. Figure 2 shows that the amplitude entropy and approximate entropy are excellent.

  The results are as follows. In the simulation data for increasing irregularity by adding random numbers in the deterministic data generation process (method 1), as shown in FIG. 1, the Shannon amplitude entropy value and the Rampelchev complexity value increase and increase linearly. Showed a trend. On the other hand, the approximate entropy showed non-linearity with a peak at a randomness of 20%. On the other hand, in the data generated by the logistic map function, the approximate entropy increased with the increase in the number of attractors as shown in Fig. 2, but the Rampelchev complexity is almost constant up to 32 attractors, Increased rapidly. From these results, it was found that good linearity and high sensitivity are required for anesthetic depth measurement. Therefore, assuming that the data generated by Method 1 is close to the electroencephalogram in the clinic, both the amplitude entropy and Rampelchev complexity methods are significant. Approximate entropy must account for limitations in use rather than exhibiting non-linearity. On the other hand, when the probabilistic data generation is assumed to be an electroencephalogram, the approximate entropy is excellent in both linearity and sensitivity.

The electroencephalogram is considered to contain linear and non-linear data. Therefore, it is considered that a method excellent in both data sets is excellent as a measurement method of electroencephalogram information and a display method and apparatus for the processing result. If the function of the data set used for the simulation is able to approximate the brain wave well, amplitude entropy and spectral entropy are the best, followed by Rampelchev complexity. Based on the above simulation results, the present invention adopts published indices of multiple algorithms with emphasis on amplitude entropy, spectral entropy, and Lampelchev complexity, and provides sedation and consciousness levels from shallow to deep, Define the range in which the efficacy of multiple anesthetics exists in multiple ranges. Next, the index ci of sedation and consciousness level is calculated by the product xi · ki of the electroencephalogram randomness xi and the weighting factor ki for each anesthetic, corresponding to the drunk drug i (i = 1,2, ... N). And an index C at the specified sedation / consciousness level is given by a linear combination Σci of each ci. That is,
C = Σci = Σxi ・ ki (i = 1,2, …… N) (2)
The logic thus obtained is converted into firmware. As a result, an electroencephalogram information measurement process corresponding to an arbitrary sedation / consciousness level from a shallow depth to a deep depth, and a display method and apparatus for the processing result are configured.

  FIG. 3 is a block diagram showing an embodiment of the electroencephalogram measurement display apparatus according to the present invention. In FIG. 3, reference numeral 1 is a first electroencephalogram state including entropy and electroencephalogram complexity such as entropy, (power) spectrum entropy, and Rampelchev complexity, which are generated as first electroencephalogram state information from a pre-measured electroencephalogram. EEG information indicating the correlation between the information index, blood drug concentration and anesthesia depth (electroencephalogram consciousness level), 2 is processed in real time by a plurality of known algorithms, and as second EEG state information An electroencephalogram information processing means for obtaining electroencephalogram entropy, electroencephalogram power spectrum, electroencephalogram complexity information, 3 refers to the electroencephalogram information, and sedation and consciousness level from shallow to deep, and a plurality of anesthesia in a plurality of ranges The range in which the efficacy of the drug exists is defined in association with the index, and multiple anesthetics and indices corresponding to the anesthetic i (i = 1,2, …… N) and the index are defined. An algorithm linear for giving a sedation / consciousness level index ci by a product xi · ki of the electroencephalogram randomness xi and weighting factor ki, and giving a specified sedation / consciousness level index C by a linear combination Σci of each ci The combination means 4 refers to the electroencephalogram information and the algorithm linear combination means, and the determination for determining the depth of anesthesia in real time as a function of the administered drug concentration from the second electroencephalogram state information obtained from the electroencephalogram information processing means Means 5 is a control means for controlling and optimizing the drug administration condition based on the determination result obtained by the determination means, and 6 is a display means for displaying the determination / control state / instruction.

  The EEG information entropy, the (power) spectrum entropy, the Rampelchev complexity, etc., and the EEG complexity information that make up the EEG information 1 are generated as the first EEG state information in advance. It is to be incorporated. Therefore, such electroencephalogram state information is generated in advance using a large number of examiners, and is collected as statistics. The electroencephalogram entropy includes approximate entropy, amplitude entropy, spectral entropy, state entropy, and reaction entropy, and is employed as an index of anesthesia depth level. The power spectrum is a classic method, divided into delta, theta, sub-alpha, alpha, and beta waves, but absolute delta power is an indicator of sedation level. On the other hand, the intermediate frequency and the spectrum edge frequency 95 can be used as indexes. The EEG complexity information includes neural complexity, KL-complexity, Ramperchev complexity, and the like. These attributes of the electroencephalogram information include the depth of anesthesia and are registered as components of the database.

  Anesthetic agents include inhalation anesthetics such as desflurane, sevoflurane and isoflurane, and intravenous anesthetics such as propofol, midazolam and fentanyl. The attributes of these anesthetics are registered in the database in association with the electroencephalogram state information.

  Processing algorithms such as approximate entropy, amplitude entropy, spectral entropy, state entropy, reaction entropy, neural complexity, KL-complexity, and Rampelchev complexity are published in many documents such as Anesthesia magazine as described above. It is possible to trace the analysis. The electroencephalogram information processing means 2 processes the measured electroencephalogram in real time using a plurality of known algorithms, and obtains electroencephalogram entropy, electroencephalogram power spectrum, electroencephalogram complexity information, etc. as second electroencephalogram state information. Algorithm linear combination means 3 refers to the above-mentioned electroencephalogram information 1 and defines sedation and consciousness levels from shallow to deep, by associating ranges where multiple anesthetics are effective in multiple ranges with indices The sedation and consciousness level is determined by the product xi · ki of the electroencephalogram randomness xi and the weight coefficient ki for each anesthetic and index corresponding to the anesthetic i (i = 1,2, …… N). The index ci at the specified sedation / consciousness level is given by a linear combination Σci of each ci. The determination means 4 refers to the first electroencephalogram state information and the algorithm linear combination means 3 registered in the above-described electroencephalogram information 1, and the first information obtained from the above-described electroencephalogram information processing means 2 and the algorithm linear combination means 3. Based on the content of the corresponding electroencephalogram information from the second electroencephalogram state information and its linear combination, the consciousness level is determined in real time as a function of the administered drug concentration.

As a result of the determination, if the current anesthetic depth or consciousness level is different from the expected value, the control means 5 controls the drug administration condition according to the determination result obtained by the determination means 4 described above, To optimize. This control needs to be performed in real time, and a loop from measurement to determination, determination to control, and control to optimization must respond quickly to the biological system.
Accordingly, the display means 6 accurately displays the above-described determination and control states and instructions.

  The electroencephalogram measurement processing display device shown in FIG. 3 is also configured as a monitor that detects the onset of sleep apnea syndrome. At this time, reference numeral 1 in FIG. 3 includes first entropy such as brain wave information entropy, (power) spectrum entropy, and Ramppelchef complexity, which are generated as first brain wave state information from previously measured brain waves, and brain wave complexity. EEG information indicating the correlation between the brain wave status information and the level of consciousness indicating EEG awakening / sleeping. 2 is processed in real time by multiple known algorithms to obtain second EEG status information. An electroencephalogram information processing means for obtaining electroencephalogram entropy, electroencephalogram power spectrum, electroencephalogram complexity information, 3 defines the sedation and consciousness level from wakefulness to sleep in association with indices in a plurality of ranges with reference to the electroencephalogram information In addition, the sedation / consciousness level is determined by the product xi · ki of the electroencephalogram randomness xi and the weighting factor ki for each of the plurality of indices corresponding to the plurality of indices. The algorithm linear combination means for giving the index ci of the sedation and the index C at the specified sedation / consciousness level by the linear combination Σci of each ci, 4 refers to the brain wave information and the algorithm linear combination means, and from the brain wave information processing means Determination means for determining the depth of anesthesia in real time as a function of the administered drug concentration from the obtained second electroencephalogram state information, 5 is optimal by controlling the drug administration conditions according to the determination result obtained by the determination means Reference numeral 6 denotes control means for converting to a display means 6 for displaying the determination and control status / instructions.

  The EEG information entropy, the (power) spectrum entropy, the Rampelchev complexity, etc., and the EEG complexity information that make up the EEG information 1 are generated as the first EEG state information in advance. It is to be incorporated. Therefore, such electroencephalogram state information is generated in advance using a large number of examiners, and is collected as statistics. The electroencephalogram entropy includes approximate entropy, amplitude entropy, spectral entropy, state entropy, and reaction entropy, and is adopted as an index of the consciousness level of the electroencephalogram. The power spectrum is a classic method, divided into delta, theta, sub-alpha, alpha, and beta waves, but absolute delta power is an indicator of sedation level. On the other hand, the intermediate frequency and the spectrum edge frequency 95 can be used as indexes. The EEG complexity information includes neural complexity, KL-complexity, Ramperchev complexity, and the like. These attributes of the electroencephalogram information are registered as database components.

  Processing algorithms such as approximate entropy, amplitude entropy, spectral entropy, state entropy, reaction entropy, neural complexity, KL-complexity, and Rampelchev complexity are published in many documents such as Anesthesia magazine as described above. It is possible to trace the analysis. The electroencephalogram information processing means 2 processes the measured electroencephalogram in real time by a plurality of known algorithms, and obtains electroencephalogram entropy, electroencephalogram power spectrum, electroencephalogram complexity information, etc. as second electroencephalogram state information. The algorithm linear combination means 3 refers to the aforementioned electroencephalogram information 1 and defines sedation and consciousness levels from awakening to sleep in association with a plurality of indicators in a plurality of ranges, and the electroencephalogram for each of the plurality of indicators is defined. An index ci of the sedation / consciousness level is given by the product xi · ki of the randomness xi and the weight coefficient ki, and an index C at the specified sedation / consciousness level is given by a linear combination Σci of each ci. The determination means 4 refers to the first electroencephalogram state information and the algorithm linear combination means 3 registered in the above-described electroencephalogram information 1, and the first information obtained from the above-described electroencephalogram information processing means 2 and the algorithm linear combination means 3. Based on the content of the corresponding electroencephalogram information, the consciousness level of wakefulness and sleep is determined in real time from the second electroencephalogram state information and its linear combination.

As a result of the determination, if the current consciousness level is different from the expected value, the control unit 5 performs control according to the determination result obtained by the determination unit 4 described above to normalize the respiratory state and change the electroencephalogram state. Normalize. This control needs to be performed in real time, and the loop from measurement to determination, determination to control, and control to normalization must respond quickly to the biological system.
Accordingly, the display means 6 accurately displays the above-described determination and control states and instructions.

  The electroencephalogram measurement and display device according to the present invention has a wide range of applications not only at the surgical operation site, but also as a consciousness level monitor due to illness or injury, a consciousness level monitor such as sleep apnea syndrome, and other health management devices.

Plot showing the result of simulating the relationship of index value to randomness as a linear quantity The result of simulating the index value for the number of branches by nonlinear dynamics. It is a block diagram which shows one Example of the electroencephalogram measurement display apparatus by this invention.

Explanation of symbols

1 EEG information 1
2 EEG information processing means 3 Algorithm linear combination means 4 Judgment means 5 Control means 6 Display means

Claims (4)

EEG information entropy generated as first EEG status information from previously measured EEG, (power) spectrum entropy, index consisting of entropy such as Rampelchev complexity and EEG complexity information, blood drug concentration and anesthetic depth ( EEG information representing the correlation with the EEG consciousness level)
EEG measured in real time is processed by multiple known algorithms to obtain EEG information entropy, power spectrum, EEG complexity information, etc. as second EEG status information,
With reference to the electroencephalogram information, the sedation and consciousness level from shallow to deep are defined in relation to the index with a range in which the efficacy of a plurality of anesthetics exists in a plurality of ranges, and the anesthetic i ( Corresponding to i = 1,2, ... N), the sedation / consciousness level index ci is given and specified by the product xi · ki of the electroencephalogram randomness xi and the weighting factor ki for each of anesthetics and indices. An index C at the sedation / consciousness level is given by a linear combination Σci of each ci,
With reference to the EEG information and the primary combination Σci of each ci, from the second EEG state information, determine the depth of anesthesia in real time as a function of the administered drug concentration,
Control to optimize drug administration conditions according to the result of the determination,
An electroencephalogram measurement and display method comprising each procedure for displaying the determination and control status. EEG information entropy generated as first EEG status information from previously measured EEG, (power) spectrum entropy, index consisting of entropy such as Rampelchev complexity and EEG complexity information, blood drug concentration and anesthetic depth ( EEG information representing the correlation with the EEG consciousness level)
An electroencephalogram information processing means for processing electroencephalograms measured in real time by a plurality of known algorithms and obtaining electroencephalogram information entropy, power spectrum, electroencephalogram complexity information, etc. as second electroencephalogram state information;
With reference to the electroencephalogram information, the sedation and consciousness level from shallow to deep are defined in relation to the index with a range in which the efficacy of a plurality of anesthetics exists in a plurality of ranges, and the anesthetic i ( Corresponding to i = 1,2, ... N), the sedation / consciousness level index ci is given and specified by the product xi · ki of the electroencephalogram randomness xi and the weighting factor ki for each of anesthetics and indices. An algorithm linear combination means for giving an index C at the sedation / consciousness level by a linear combination Σci of each ci,
Determining means for determining in real time the depth of anesthesia as a function of the administered drug concentration from the second electroencephalogram state information obtained from the electroencephalogram information processing means with reference to the electroencephalogram information and the algorithm linear combination means; ,
Control means for optimizing drug administration conditions according to the determination result obtained by the determination means;
An electroencephalogram measurement and display device comprising display means for displaying the determination and control status. EEG information entropy generated as first EEG state information from previously measured EEG, (power) spectrum entropy, Ramperchev complexity, etc. Get brain wave information,
EEG measured in real time is processed by multiple known algorithms to obtain EEG information entropy, power spectrum, EEG complexity information, etc. as second EEG status information,
With reference to the electroencephalogram information, sedation and consciousness levels from awakening to sleep are defined in association with a plurality of the indicators in a plurality of ranges, and the electroencephalogram randomness xi and the weight coefficient ki for each of the plurality of indicators An index ci of the sedation / consciousness level is given by the product xi · ki, and an index C at the specified sedation / consciousness level is given by a linear combination Σci of each ci,
With reference to the brain wave information and the primary combination Σci of each ci, from the second brain wave state information, determine sedation and consciousness level in real time,
Control to normalize sedation and consciousness level according to the result of the determination,
An electroencephalogram measurement and display method comprising each procedure for displaying the determination and control status. EEG information entropy generated as first EEG state information from previously measured EEG, (power) spectrum entropy, Ramperchev complexity, etc. EEG information and
An electroencephalogram information processing means for processing electroencephalograms measured in real time by a plurality of known algorithms and obtaining electroencephalogram information entropy, power spectrum, electroencephalogram complexity information, etc. as second electroencephalogram state information;
With reference to the electroencephalogram information, sedation and consciousness levels from awakening to sleep are defined in association with a plurality of the indicators in a plurality of ranges, and the electroencephalogram randomness xi and the weight coefficient ki for each of the plurality of indicators An algorithm linear combination means for giving an index ci of a sedation / consciousness level by a product xi · ki and giving an index C at a specified sedation / consciousness level by a linear combination Σci of each ci,
With reference to the electroencephalogram information and the algorithm linear combination means, from the second electroencephalogram state information obtained from the electroencephalogram information processing means, a determination means for determining sedation and consciousness level in real time;
Control means for normalizing sedation and consciousness level according to the determination result obtained by the determination means;
Display means for displaying the determination and control status;
An electroencephalogram measurement display device comprising:

Images