CN117643474B - Electroencephalogram signal processing method and system based on deep neural network - Google Patents
Electroencephalogram signal processing method and system based on deep neural network Download PDFInfo
- Publication number
- CN117643474B CN117643474B CN202311784847.9A CN202311784847A CN117643474B CN 117643474 B CN117643474 B CN 117643474B CN 202311784847 A CN202311784847 A CN 202311784847A CN 117643474 B CN117643474 B CN 117643474B
- Authority
- CN
- China
- Prior art keywords
- voltage
- brain wave
- judgment
- frequency
- outputting
- 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.)
- Active
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 25
- 238000003672 processing method Methods 0.000 title claims abstract description 11
- 210000004556 brain Anatomy 0.000 claims abstract description 143
- 238000012545 processing Methods 0.000 claims abstract description 59
- 238000010586 diagram Methods 0.000 claims abstract description 44
- 238000004458 analytical method Methods 0.000 claims abstract description 43
- 238000001914 filtration Methods 0.000 claims abstract description 39
- 238000003379 elimination reaction Methods 0.000 claims abstract description 37
- 230000008030 elimination Effects 0.000 claims abstract description 31
- 238000000034 method Methods 0.000 claims abstract description 13
- 238000012790 confirmation Methods 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 17
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 description 6
- 230000003183 myoelectrical effect Effects 0.000 description 5
- 230000004424 eye movement Effects 0.000 description 4
- 238000013019 agitation Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000002996 emotional effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000002238 attenuated effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000004397 blinking Effects 0.000 description 1
- 230000007177 brain activity Effects 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 208000035475 disorder Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002565 electrocardiography Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 208000028327 extreme fatigue Diseases 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 210000004761 scalp Anatomy 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 230000002618 waking effect Effects 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/369—Electroencephalography [EEG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/369—Electroencephalography [EEG]
- A61B5/372—Analysis of electroencephalograms
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Surgery (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Pathology (AREA)
- Public Health (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Heart & Thoracic Surgery (AREA)
- Veterinary Medicine (AREA)
- Biomedical Technology (AREA)
- Psychiatry (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Signal Processing (AREA)
- Theoretical Computer Science (AREA)
- Psychology (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses an electroencephalogram signal processing method and system based on a deep neural network, and relates to the technical field of electroencephalogram signal processing, wherein the method comprises the steps of acquiring an electroencephalogram signal diagram, performing error elimination processing on the electroencephalogram signal diagram, and outputting an error elimination electroencephalogram diagram; setting the filtering frequency of a band-pass filter, filtering the error brain wave image, and outputting a filtering signal image; dividing the filtered signal diagram, and outputting dividing time; the method is used for solving the problem that in the prior art, the analysis of the voltage with larger amplitude in the electroencephalogram is lack, so that the voltage is still interfered by artifacts in the subsequent analysis, and the accuracy of the analysis is influenced when the electroencephalogram is analyzed.
Description
Technical Field
The invention relates to the technical field of electroencephalogram signal processing, in particular to an electroencephalogram signal processing method and system based on a deep neural network.
Background
The electroencephalogram signal processing technology refers to a series of methods and technologies for analyzing and processing electroencephalogram signals; the brain electrical signal is an electrophysiological signal for recording brain activity, and reflects the electrical activity of brain neurons by measuring the potential change on the scalp; electroencephalogram signal processing technology aims at extracting useful information from original electroencephalogram data so as to achieve different research targets and applications.
The existing improvement for electroencephalogram signal processing is usually to optimize an electroencephalogram signal acquisition device and a filtering device, so that the number and the size of the electroencephalogram signal acquisition device are reduced, for example, in China patent with the application publication number CN112949586A, an electroencephalogram signal processing chip and an electroencephalogram signal processing system are disclosed.
Disclosure of Invention
The invention aims to solve at least one of the technical problems in the prior art to a certain extent, and aims to solve the problem that the accuracy of analysis is affected when the electroencephalogram is analyzed due to the fact that the follow-up analysis is still interfered by artifacts due to the fact that the analysis of the voltage with larger amplitude in the electroencephalogram is lack in the prior art by improving the filtering technology of the electroencephalogram processing.
To achieve the above object, in a first aspect, the present invention provides an electroencephalogram signal processing method based on a deep neural network, including:
Step S1, acquiring an electroencephalogram in a first time, wherein the electroencephalogram comprises voltage and acquisition time; analyzing the voltage and the acquisition time, performing error elimination processing on the electroencephalogram based on the analysis result, and outputting an error elimination electroencephalogram; analyzing the brain wave diagram, and outputting brain wave number information based on the analysis result;
step S2, setting the filtering frequency of a band-pass filter based on the brain wave number information, filtering the error brain wave image by using the band-pass filter, and outputting a filtering signal image;
step S3, dividing the filtered signal diagram based on the brain wave number information and the first time, and outputting dividing time;
s4, analyzing and calculating the voltage of the filtered signal diagram in the dividing time, and obtaining a reference judgment frequency based on a calculation result; and analyzing the reference judgment frequency, and outputting brain wave specific information based on an analysis result.
Further, the step S1 includes the following sub-steps:
step S1011, acquiring a continuous electroencephalogram in a first time, wherein the ordinate of the electroencephalogram is voltage, and the abscissa is acquisition time;
step S1012, obtaining the maximum value of the voltage, and marking the maximum value as the maximum voltage;
Step S1013, calculating the absolute value of the difference value between two groups of voltages adjacent to each other in the acquisition time sequence, wherein the absolute value is marked as an nth peak judgment voltage, n is the peak judgment voltage number, and n is a positive integer;
step S1014, calculating by using a jump judgment formula to obtain a jump judgment value;
the jump judgment formula is configured as follows: ju= ; Wherein JU is jump judgment value, PV is peak judgment voltage, and Max is maximum voltage;
In step S1015, a peak judgment voltage with a positive jump judgment value is obtained, and the maximum value of the two sets of voltages for calculating the peak judgment voltage is marked as a confirmed jump value.
Further, the error elimination process includes an average voltage process and an artifact elimination process, and the step S1 further includes the following sub-steps:
step S1021, calculating the duty ratio of the jump value to the total voltage number in the electroencephalogram, and marking the duty ratio as an artifact judgment ratio;
step S1022, when the artifact judgment value is smaller than the artifact threshold value, carrying out average voltage processing on the confirmed jump value;
the average voltage processing comprises the steps of obtaining a voltage adjacent to the left and a voltage adjacent to the right of the obtaining time of the confirmation jump value, calculating the average value of the voltage adjacent to the left and the voltage adjacent to the right, and updating the confirmation jump value to the average value of the voltage adjacent to the left and the voltage adjacent to the right;
Step S1023, when the artifact judgment value is greater than or equal to the artifact threshold value, artifact removal processing is performed.
Further, the artifact removal processing includes:
Sequencing the confirmed jump values from front to back according to the acquisition time to obtain a jump value sequence;
calculating to obtain the occurrence interval of the adjacent confirmation jump value by using the elimination time formula;
The cancellation time formula is configured to: TM i＝GNi+1-GNi; wherein TM is the occurrence interval, GN is the acquisition time for confirming the jump value, i is the occurrence interval and the number of the acquisition time, and i is a positive integer;
calculating the average value of all the occurrence intervals, and marking the average value as an average interval;
analyzing the average interval by using a deep neural network, and outputting artifact information;
Performing artifact elimination processing on the electroencephalogram by using a deep neural network based on artifact information, and outputting an error-elimination electroencephalogram, wherein the error-elimination electroencephalogram comprises voltages;
And obtaining the maximum value of the voltage in the error-eliminating brain wave diagram, and updating the maximum voltage to the maximum value of the voltage in the error-eliminating brain wave diagram.
Further, the brain wave number amount information includes first brain wave number amount information, second brain wave number amount information, and third brain wave number amount information, and the step S1 further includes the steps of:
Step S103, analyzing the maximum voltage, and outputting first brain wave number information when the maximum voltage is greater than or equal to a first judgment voltage;
Outputting second brain wave number information when the maximum voltage is greater than the first judgment voltage and less than or equal to the second judgment voltage;
And outputting the third brain wave number information when the maximum voltage is larger than the second judgment voltage and smaller than the third judgment voltage.
Further, the step S2 includes the following sub-steps:
step S201, receiving brain wave number information, setting the low-pass frequency of the band-pass filter to be a first low frequency and setting the high-pass frequency of the band-pass filter to be a first high frequency when receiving the first brain wave number information;
Step S202, when the second brain wave number information or the third brain wave number information is received, setting the low-pass frequency of the band-pass filter to be a second low frequency and setting the high-pass frequency of the band-pass filter to be a second high frequency;
step S203, the electroencephalogram signal is filtered by a band-pass filter, and a filtered signal diagram is output.
Further, the step S3 includes the following sub-steps:
step S301, when the first brain wave number is received, setting the dividing time as a second time;
step S302, when the second brain wave number is received, setting the dividing time as a third time;
step S303, when the third brain wave number is received, setting the dividing time as a fourth time;
step S304, dividing the first time into a plurality of time periods with equal or unequal duration according to the dividing time, and outputting the dividing time.
Further, the reference judgment frequency includes a first judgment frequency and a second judgment frequency, and the step S4 includes the following sub-steps:
step S401, setting a first calculation multiple of the second judgment voltage as a first calculation voltage; setting a second calculation multiple of the second judgment voltage as a second calculation voltage;
step S402, calculating the occurrence frequency of the first calculated voltage and the second calculated voltage in the dividing time, wherein the occurrence frequency is respectively marked as a first judgment frequency and a second judgment frequency;
step S403, analyzing the first judgment frequency and the second judgment frequency, and outputting brain wave specific information based on the analysis result; the brain wave specific information comprises alpha brain wave information, beta brain wave information, gamma brain wave information, delta brain wave information and theta brain wave information.
Further, the step S403 includes:
When the first judgment frequency and the second judgment frequency are in the first frequency interval, delta brain wave information is output;
Outputting theta brain wave information when the first judgment frequency is in the second frequency interval and the second judgment frequency is in the first or second frequency interval;
Outputting alpha brain wave information when the second judgment frequency is in a third frequency interval;
Outputting beta brain wave information when the first judgment frequency is in a fourth frequency interval and the second judgment frequency is in the first frequency interval;
and outputting gamma brain wave information when the first judgment frequency is larger than the fourth frequency interval and the second judgment frequency is in the first frequency interval.
In a second aspect, the invention also provides an electroencephalogram signal processing system based on the deep neural network, which comprises an error processing module, a voltage processing module and an electroencephalogram analysis module; the error processing module comprises an error confirmation unit and an error elimination unit; the error confirmation unit is used for calculating peak value judgment voltage and judging whether a confirmation jump value exists or not based on the peak value judgment voltage; the error elimination unit is used for analyzing the confirmed jump value, carrying out error elimination processing on the confirmed jump value based on an analysis result and outputting an error elimination brain wave diagram;
The voltage processing module comprises a filtering analysis unit and a time dividing unit; the filtering analysis unit is used for analyzing the voltage after the error processing and outputting brain wave number information; setting the filtering frequency of a band-pass filter based on brain wave number information, filtering a message brain wave diagram, and outputting a filtering signal diagram; the time dividing unit is used for dividing the filtered signal diagram according to the brain wave number information and outputting dividing time;
The brain wave analysis module is used for calculating a reference judgment frequency, analyzing the reference judgment frequency and outputting brain wave specific information based on an analysis result.
The invention has the beneficial effects that: according to the brain wave analysis method, the maximum voltage of the brain wave signal graph is analyzed, based on the analysis result, the brain wave number information can be output, and different filtering frequencies are set according to the brain wave number information, so that the brain wave analysis method has the advantages that accurate filtering signal graphs under different brain waves can be obtained, physiological artifacts and non-physiological artifacts can be better filtered, the subsequent analysis of the voltage of the filtering signal graph is facilitated, and the accuracy of brain wave signal analysis is improved;
The brain wave specific information is output based on the reference judgment frequency by setting different division times, calculating the reference judgment frequency of the specific voltage in the division times, and the brain wave specific information processing method has the advantages that different brain waves cannot suddenly change in a certain time under normal conditions, and whether the brain electric signal is in a stable state can be judged by setting different division times, so that the intelligence of the brain electric signal processing can be improved.
Additional features and advantages of the application will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the application. The objectives and other advantages of the application will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.
Drawings
FIG. 1 is a process step diagram of the present invention;
FIG. 2 is a schematic diagram of an artifact signal according to the present invention;
FIG. 3 is an electroencephalogram diagram of the present invention;
FIG. 4 is a diagram of a filtered signal according to the present invention;
Fig. 5 is a system schematic block diagram of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
Referring to fig. 1, the invention provides an electroencephalogram signal processing method based on a deep neural network, which can acquire an electroencephalogram signal diagram and analyze and output brain wave number information; setting a filtering frequency, performing filtering treatment, and outputting a filtering signal diagram; dividing the filtered signal diagram, and outputting dividing time; calculating reference judgment frequency, analyzing and outputting brain wave specific information
Specifically, the method comprises the following steps:
Step S1, acquiring an electroencephalogram in a first time, wherein the electroencephalogram comprises voltage and acquisition time; analyzing the voltage and the acquisition time, performing error elimination processing on the electroencephalogram based on the analysis result, and outputting an error elimination electroencephalogram; analyzing the brain wave diagram, and outputting brain wave number information based on the analysis result; step S1 further comprises the following sub-steps:
Step S1011, acquiring a continuous electroencephalogram in a first time, wherein the ordinate of the electroencephalogram is voltage, and the abscissa of the electroencephalogram is acquisition time;
in practice, the first time is set to one minute;
step S1012, obtaining the maximum value of the voltage, and marking the maximum value as the maximum voltage;
Step S1013, calculating the absolute value of the difference value between two groups of voltages adjacent to each other in the acquisition time sequence, wherein the absolute value is marked as an nth peak judgment voltage, n is the peak judgment voltage number, and n is a positive integer;
step S1014, calculating by using a jump judgment formula to obtain a jump judgment value;
the jump judgment formula is configured as follows: ju= ; Wherein JU is jump judgment value, PV is peak judgment voltage, and Max is maximum voltage;
Step S1015, obtaining a peak value judgment voltage with a positive jump judgment value, and marking the maximum value of two groups of voltages for calculating the peak value judgment voltage as a confirmed jump value;
referring to fig. 2, in fig. 2, T1 is voltage fluctuation caused by blinking, T2 is voltage fluctuation caused by myoelectricity, and T3 is voltage fluctuation caused by electrocardiography;
step S1021, calculating the duty ratio of the jump value to the total voltage number in the electroencephalogram, and marking the duty ratio as an artifact judgment ratio;
step S1022, when the artifact judgment value is smaller than the artifact threshold value, carrying out average voltage processing on the confirmed jump value;
In practice, the artifact threshold is set to 5%;
it should be noted that, setting the artifact threshold is related to the sampling frequency of the electroencephalogram signal, if the sampling frequency is higher, the artifact threshold is lower in practical application;
the average voltage processing includes acquiring a voltage adjacent to the left and a voltage adjacent to the right of the acquisition time of the confirmation jump value, calculating an average value of the voltage adjacent to the left and the voltage adjacent to the right, and updating the confirmation jump value to the average value of the voltage adjacent to the left and the voltage adjacent to the right;
Since the electroencephalogram signal is continuous, the confirmation jump value cannot be directly deleted, and therefore, the average value of the voltages adjacent to each other around the confirmation jump value is selected and calculated as the voltage of the point;
step S1023, when the artifact judgment value is greater than or equal to the artifact threshold value, artifact elimination processing is performed;
the artifact removal process includes:
Sequencing the confirmed jump values from front to back according to the acquisition time to obtain a jump value sequence;
calculating to obtain the occurrence interval of the adjacent confirmation jump value by using the elimination time formula;
The cancellation time formula is configured to: tmi= GNi +1-GNi; wherein TM is the occurrence interval, GN is the acquisition time for confirming the jump value, i is the occurrence interval and the number of the acquisition time, and i is a positive integer;
calculating the average value of all the occurrence intervals, and marking the average value as an average interval;
analyzing the average interval by using a deep neural network, and outputting artifact information;
It should be noted that, the artifact information may include an electrocardiographic artifact, an eye movement artifact, and an myoelectric artifact, and the artifact is unavoidable when acquiring the electroencephalogram signal, and the electrocardiographic artifact is shown in fig. 2 to appear regularly, and the eye movement artifact and the myoelectric artifact may appear in a small amount and disorder compared with the electrocardiographic artifact; the eye movement artifact and the myoelectric artifact can be identified by setting the artifact threshold; carrying out average voltage processing on the eye movement artifacts, carrying out artifact elimination processing on the electrocardio artifacts by using a deep neural network technology, and removing the myoelectricity artifacts by subsequent filtering processing;
performing artifact elimination processing on the electroencephalogram by using a deep neural network based on artifact information, and outputting an error elimination electroencephalogram which comprises voltages;
The deep neural network is an artifact identification model obtained by training big data by using a deep neural network technology, and the model can identify voltage fluctuation caused by electrocardio artifacts in an electroencephalogram and eliminate artifact fluctuation parts;
obtaining the maximum value of the voltage in the error-eliminating brain wave diagram, and updating the maximum voltage to the maximum value of the voltage in the error-eliminating brain wave diagram;
Step S103, analyzing the maximum voltage: outputting first brain wave number information when the maximum voltage is greater than or equal to the first judgment voltage;
Outputting second brain wave number information when the maximum voltage is greater than the first judgment voltage and less than or equal to the second judgment voltage;
outputting third brain wave number information when the maximum voltage is larger than the second judgment voltage and smaller than the third judgment voltage;
In specific implementation, the first judgment voltage is set to 30 μv, the second judgment voltage is set to 50 μv, and the third judgment voltage is set to 100 μv;
The brain waves include delta waves of 10-30 mu V, which are generated by an adult under the condition of extreme fatigue or deep sleep, and the delta waves are not more than 50 mu V; the theta wave of the adult under the shallow sleep state is 10-30 mu V; alpha waves of 30-50 mu V, which are generated by adults in a conscious and quiet state, are not more than 100 mu V at the highest; beta waves of 5-30 mu V, generally 20 mu V, appear under the condition that the attention of an adult is focused; the gamma wave appears at 15-25 mu V when the adult is in stress and emotional agitation or excitation; five possible waveforms;
Setting the first judgment voltage to 30 μv, when the maximum voltage is less than or equal to 30 μv, since δ and θ waves can maximally reach 50 μv in the normal range, and no voltage greater than 30 μv is present in the electroencephalogram, it cannot be judged whether δ and θ waves are present, and therefore the first brain wave number information is output indicating that β waves or γ waves may be present;
Setting the second judgment voltage to 50 μv, when the maximum voltage is less than or equal to 50 μv, the same cannot accurately judge whether or not the α wave exists, and therefore the second brain wave number information is output to indicate that the δ, θ, β, and γ waves may exist;
Setting the third judgment voltage to 100 mu v, when the maximum voltage is greater than 50 mu v and less than 100 mu v, accurately judging that alpha waves exist at the moment, and outputting third brain wave number information which indicates that alpha, delta, theta, beta and gamma waves possibly exist;
Referring to fig. 3 and 4, fig. 4 is a diagram of a filtered signal obtained by filtering the signal of fig. 3 with a low-pass frequency of 40Hz and a high-pass frequency of 1 Hz;
Step S2, setting the filtering frequency of a band-pass filter based on the brain wave number information, filtering the error brain wave image by using the band-pass filter, and outputting a filtering signal image; step S2 further comprises the following sub-steps:
step S201, receiving brain wave number information, setting the low-pass frequency of the band-pass filter to be a first low frequency and setting the high-pass frequency of the band-pass filter to be a first high frequency when receiving the first brain wave number information;
Step S202, when the second brain wave number information or the third brain wave number information is received, setting the low-pass frequency of the band-pass filter to be a second low frequency and setting the high-pass frequency of the band-pass filter to be a second high frequency;
Step S203, filtering the electroencephalogram by using a band-pass filter, and outputting a filtered signal diagram;
in a specific implementation, the first low frequency is set to 50Hz, the second low frequency is set to 40Hz, the first high frequency is set to 0.5Hz, and the second high frequency is set to 0.1Hz;
the waveform frequency domains of the five brain waves are respectively 0.5-3 Hz of delta wave; theta wave 4-8 Hz; alpha wave 8-12 Hz; beta wave 13-32 Hz; gamma wave 32-44 Hz;
High-pass filtering at 0.5Hz and 0.1Hz can attenuate skin potential or other slow voltages, so that the waveform is smoother; line noise and myoelectric noise can be attenuated with a low pass filter of 40 Hz; however, when the low-pass filter is set to 40Hz, accurate analysis of the gamma wave cannot be performed, so the low-pass frequency is set to 50Hz; at this time, since the delta wave does not exist, the high-pass frequency is set to 0.5Hz to make the waveform smoother;
When the second brain wave number information or the third brain wave number information is received, the delta wave is indicated to be possibly present, and the high-pass frequency is set to be 0.1Hz at the moment, so that the delta wave can be accurately analyzed; the low-pass frequency is set to be 40Hz, so that line noise and myoelectric noise can be reduced, and analysis of four waveforms except gamma waves is facilitated;
Step S3, dividing the filtered signal diagram based on the brain wave number information and the first time, and outputting dividing time; step S3 further comprises the following sub-steps:
step S301, when the first brain wave number is received, setting the dividing time as a second time;
step S302, when the second brain wave number is received, setting the dividing time as a third time;
step S303, when the third brain wave number is received, setting the dividing time as a fourth time;
step S304, dividing the first time into a plurality of time periods with equal or unequal duration according to the dividing time, and outputting the dividing time;
In practice, the second time is set to 5s, and the adult may be in a beta wave awake or gamma wave emotional agitation state, and the time for transition from the awake or awake agitation state to other states is usually not less than 5s; the third time is set to 30s, and at the moment, an adult can be in a delta wave deep sleep state or a theta wave shallow sleep state, and under normal conditions, the adult can keep 30s unchanged without human factor interference; the fourth time is set to 1s, and at this time, an adult may be in an alpha wave initial awake state, which exists only during the period from just waking up to opening eyes, and the dividing time is set to 1s because of the short time;
s4, analyzing and calculating the voltage of the filtered signal diagram in the dividing time, and obtaining a reference judgment frequency based on a calculation result; analyzing the reference judgment frequency, and outputting brain wave specific information based on an analysis result; step S4 further comprises the sub-steps of:
step S401, setting a first calculation multiple of the second judgment voltage as a first calculation voltage; setting a second calculation multiple of the second judgment voltage as a second calculation voltage;
step S402, calculating the occurrence frequency of the first calculated voltage and the second calculated voltage in the dividing time, wherein the occurrence frequency is respectively marked as a first judgment frequency and a second judgment frequency;
in specific implementation, the first calculation multiple is set to 0.4, and the second calculation multiple is set to 0.8;
At this time, the first calculated voltage is 20 μv, and the second calculated voltage is 40 μv;
step S403, analyzing the first judgment frequency and the second judgment frequency, and outputting brain wave specific information based on the analysis result; the brain wave specific information comprises alpha brain wave information, beta brain wave information, gamma brain wave information, delta brain wave information and theta brain wave information;
step S403 includes:
When the first judgment frequency and the second judgment frequency are in the first frequency interval, delta brain wave information is output;
Outputting theta brain wave information when the first judgment frequency is in the second frequency interval and the second judgment frequency is in the first or second frequency interval;
Outputting alpha brain wave information when the second judgment frequency is in a third frequency interval;
Outputting beta brain wave information when the first judgment frequency is in a fourth frequency interval and the second judgment frequency is in the first frequency interval;
outputting gamma brain wave information when the first judgment frequency is larger than the fourth frequency interval and the second judgment frequency is in the first frequency interval;
in specific implementation, the first frequency interval is set to be 0-4 Hz, the second frequency interval is set to be 4-8 Hz, the third frequency interval is set to be 8-12 Hz, and the fourth frequency interval is set to be 13-32 Hz;
Example 2
Referring to fig. 3, in a second aspect, the present invention further provides an electroencephalogram signal processing system based on a deep neural network, which includes an error processing module, a voltage processing module, and an electroencephalogram analysis module; the error processing module comprises an error confirmation unit and an error elimination unit; the error confirming unit is used for calculating peak value judging voltage and judging whether a confirmed jump value exists or not based on the peak value judging voltage; the error elimination unit is used for analyzing the confirmed jump value, carrying out error elimination processing on the confirmed jump value based on the analysis result and outputting an error elimination brain wave diagram;
The voltage processing module comprises a filtering analysis unit and a time dividing unit; the filtering analysis unit is used for analyzing the voltage after the error processing and outputting brain wave number information; setting the filtering frequency of a band-pass filter based on brain wave number information, filtering a message brain wave diagram, and outputting a filtering signal diagram; the time dividing unit is used for dividing the filtered signal diagram according to the brain wave number information and outputting dividing time;
The brain wave analysis module is used for calculating a reference judgment frequency, analyzing the reference judgment frequency and outputting brain wave specific information based on an analysis result.
In the foregoing embodiments of the present application, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media having computer-usable program code embodied therein. The storage medium may be implemented by any type of volatile or non-volatile Memory device or combination thereof, such as static random access Memory (Static Random Access Memory, SRAM), electrically erasable Programmable Read-Only Memory (ELECTRICALLY ERASABLE PROGRAMMABLE READ-Only Memory, EEPROM), erasable Programmable Read-Only Memory (Erasable Programmable Read Only Memory, EPROM), programmable Read-Only Memory (PROM), read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, for example, the division of the units is merely a logical function division, and there may be other manners of division in actual implementation, and for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, device or unit indirect coupling or communication connection, which may be in electrical, mechanical or other form.
Claims (8)
1. The electroencephalogram signal processing method based on the deep neural network is characterized by comprising the following steps of:
Step S1, acquiring an electroencephalogram in a first time, wherein the electroencephalogram comprises voltage and acquisition time; analyzing the voltage and the acquisition time, performing error elimination processing on the electroencephalogram based on the analysis result, and outputting an error elimination electroencephalogram; analyzing the brain wave diagram, and outputting brain wave number information based on the analysis result; the error elimination processing comprises average voltage processing and artifact elimination processing; the brain wave number information comprises first brain wave number information, second brain wave number information and third brain wave number information, and the step S1 further comprises the following steps:
Step S103, analyzing the maximum voltage, and outputting first brain wave number information when the maximum voltage is greater than or equal to a first judgment voltage;
Outputting second brain wave number information when the maximum voltage is greater than the first judgment voltage and less than or equal to the second judgment voltage;
outputting third brain wave number information when the maximum voltage is larger than the second judgment voltage and smaller than the third judgment voltage;
step S2, setting the filtering frequency of a band-pass filter based on the brain wave number information, filtering the error brain wave image by using the band-pass filter, and outputting a filtering signal image;
step S3, dividing the filtered signal diagram based on the brain wave number information and the first time, and outputting dividing time;
S4, analyzing and calculating the voltage of the filtered signal diagram in the dividing time, and obtaining a reference judgment frequency based on a calculation result; analyzing the reference judgment frequency, and outputting brain wave specific information based on an analysis result; the reference judgment frequency comprises a first judgment frequency and a second judgment frequency, and the step S4 comprises the following sub-steps:
step S401, setting a first calculation multiple of the second judgment voltage as a first calculation voltage; setting a second calculation multiple of the second judgment voltage as a second calculation voltage;
step S402, calculating the occurrence frequency of the first calculated voltage and the second calculated voltage in the dividing time, wherein the occurrence frequency is respectively marked as a first judgment frequency and a second judgment frequency;
step S403, analyzing the first judgment frequency and the second judgment frequency, and outputting brain wave specific information based on the analysis result; the brain wave specific information comprises alpha brain wave information, beta brain wave information, gamma brain wave information, delta brain wave information and theta brain wave information.
2. The method for processing electroencephalogram signals based on the deep neural network according to claim 1, wherein the step S1 comprises the following sub-steps:
step S1011, acquiring a continuous electroencephalogram in a first time, wherein the ordinate of the electroencephalogram is voltage, and the abscissa is acquisition time;
step S1012, obtaining the maximum value of the voltage, and marking the maximum value as the maximum voltage;
Step S1013, calculating the absolute value of the difference value between two groups of voltages adjacent to each other in the acquisition time sequence, wherein the absolute value is marked as an nth peak judgment voltage, n is the peak judgment voltage number, and n is a positive integer;
step S1014, calculating by using a jump judgment formula to obtain a jump judgment value;
the jump judgment formula is configured as follows: ju= ; Wherein JU is jump judgment value, PV is peak judgment voltage, and Max is maximum voltage;
In step S1015, a peak judgment voltage with a positive jump judgment value is obtained, and the maximum value of the two sets of voltages for calculating the peak judgment voltage is marked as a confirmed jump value.
3. The method for processing an electroencephalogram signal based on a deep neural network according to claim 2, wherein the step S1 further comprises the sub-steps of:
step S1021, calculating the duty ratio of the jump value to the total voltage number in the electroencephalogram, and marking the duty ratio as an artifact judgment ratio;
step S1022, when the artifact judgment value is smaller than the artifact threshold value, carrying out average voltage processing on the confirmed jump value;
the average voltage processing comprises the steps of obtaining a voltage adjacent to the left and a voltage adjacent to the right of the obtaining time of the confirmation jump value, calculating the average value of the voltage adjacent to the left and the voltage adjacent to the right, and updating the confirmation jump value to the average value of the voltage adjacent to the left and the voltage adjacent to the right;
Step S1023, when the artifact judgment value is greater than or equal to the artifact threshold value, artifact removal processing is performed.
4. The deep neural network-based electroencephalogram signal processing method according to claim 3, wherein the artifact removal processing includes:
Sequencing the confirmed jump values from front to back according to the acquisition time to obtain a jump value sequence;
calculating to obtain the occurrence interval of the adjacent confirmation jump value by using the elimination time formula;
The cancellation time formula is configured to: TM i＝GNi+1-GNi; wherein TM is the occurrence interval, GN is the acquisition time for confirming the jump value, i is the occurrence interval and the number of the acquisition time, and i is a positive integer;
calculating the average value of all the occurrence intervals, and marking the average value as an average interval;
analyzing the average interval by using a deep neural network, and outputting artifact information;
Performing artifact elimination processing on the electroencephalogram by using a deep neural network based on artifact information, and outputting an error-elimination electroencephalogram, wherein the error-elimination electroencephalogram comprises voltages;
And obtaining the maximum value of the voltage in the error-eliminating brain wave diagram, and updating the maximum voltage to the maximum value of the voltage in the error-eliminating brain wave diagram.
5. The method for processing electroencephalogram signals based on the deep neural network according to claim 4, wherein the step S2 comprises the following sub-steps:
step S201, receiving brain wave number information, setting the low-pass frequency of the band-pass filter to be a first low frequency and setting the high-pass frequency of the band-pass filter to be a first high frequency when receiving the first brain wave number information;
Step S202, when the second brain wave number information or the third brain wave number information is received, setting the low-pass frequency of the band-pass filter to be a second low frequency and setting the high-pass frequency of the band-pass filter to be a second high frequency;
step S203, the electroencephalogram signal is filtered by a band-pass filter, and a filtered signal diagram is output.
6. The method for processing electroencephalogram signals based on the deep neural network according to claim 5, wherein the step S3 comprises the following sub-steps:
step S301, when the first brain wave number is received, setting the dividing time as a second time;
step S302, when the second brain wave number is received, setting the dividing time as a third time;
step S303, when the third brain wave number is received, setting the dividing time as a fourth time;
step S304, dividing the first time into a plurality of time periods with equal or unequal duration according to the dividing time, and outputting the dividing time.
7. The deep neural network-based electroencephalogram signal processing method according to claim 6, wherein the step S403 includes:
When the first judgment frequency and the second judgment frequency are in the first frequency interval, delta brain wave information is output;
Outputting theta brain wave information when the first judgment frequency is in the second frequency interval and the second judgment frequency is in the first or second frequency interval;
Outputting alpha brain wave information when the second judgment frequency is in a third frequency interval;
Outputting beta brain wave information when the first judgment frequency is in a fourth frequency interval and the second judgment frequency is in the first frequency interval;
and outputting gamma brain wave information when the first judgment frequency is larger than the fourth frequency interval and the second judgment frequency is in the first frequency interval.
8. A system suitable for the deep neural network-based electroencephalogram signal processing method as claimed in any one of claims 1 to 7, and comprising an error processing module, a voltage processing module and an electroencephalogram analysis module; the error processing module comprises an error confirmation unit and an error elimination unit; the error confirmation unit is used for calculating peak value judgment voltage and judging whether a confirmation jump value exists or not based on the peak value judgment voltage; the error elimination unit is used for analyzing the confirmed jump value, carrying out error elimination processing on the confirmed jump value based on an analysis result and outputting an error elimination brain wave diagram;
The voltage processing module comprises a filtering analysis unit and a time dividing unit; the filtering analysis unit is used for analyzing the voltage after the error processing and outputting brain wave number information; setting the filtering frequency of a band-pass filter based on brain wave number information, filtering a message brain wave diagram, and outputting a filtering signal diagram; the time dividing unit is used for dividing the filtered signal diagram according to the brain wave number information and outputting dividing time;
The brain wave analysis module is used for calculating a reference judgment frequency, analyzing the reference judgment frequency and outputting brain wave specific information based on an analysis result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311784847.9A CN117643474B (en) | 2023-12-24 | 2023-12-24 | Electroencephalogram signal processing method and system based on deep neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311784847.9A CN117643474B (en) | 2023-12-24 | 2023-12-24 | Electroencephalogram signal processing method and system based on deep neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117643474A CN117643474A (en) | 2024-03-05 |
| CN117643474B true CN117643474B (en) | 2024-05-28 |
Family
ID=90049507
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311784847.9A Active CN117643474B (en) | 2023-12-24 | 2023-12-24 | Electroencephalogram signal processing method and system based on deep neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117643474B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2770008Y (en) * | 2005-03-04 | 2006-04-05 | 香港理工大学 | Dozing detection alarm |
| CN106175698A (en) * | 2016-09-21 | 2016-12-07 | 广州视源电子科技股份有限公司 | Sleep cycle detection device in sleep state analysis |
| CN111012341A (en) * | 2020-01-08 | 2020-04-17 | 东南大学 | Artifact removal and electroencephalogram signal quality evaluation method based on wearable electroencephalogram equipment |
| WO2020080354A1 (en) * | 2018-10-15 | 2020-04-23 | 田辺三菱製薬株式会社 | Electroencephalogram analysis apparatus, electroencephalogram analysis system, and electroencephalogram analysis program |
-
2023
- 2023-12-24 CN CN202311784847.9A patent/CN117643474B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2770008Y (en) * | 2005-03-04 | 2006-04-05 | 香港理工大学 | Dozing detection alarm |
| CN106175698A (en) * | 2016-09-21 | 2016-12-07 | 广州视源电子科技股份有限公司 | Sleep cycle detection device in sleep state analysis |
| WO2020080354A1 (en) * | 2018-10-15 | 2020-04-23 | 田辺三菱製薬株式会社 | Electroencephalogram analysis apparatus, electroencephalogram analysis system, and electroencephalogram analysis program |
| CN111012341A (en) * | 2020-01-08 | 2020-04-17 | 东南大学 | Artifact removal and electroencephalogram signal quality evaluation method based on wearable electroencephalogram equipment |
Non-Patent Citations (1)
| Title |
|---|
| 基于CEEMDAN-ICA 的单通道脑电信号眼电伪迹滤除方法;罗志增 等;传感技术学报;20180831;第31卷(第8期);1211-1216 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117643474A (en) | 2024-03-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Glover et al. | Context-based automated detection of epileptogenic sharp transients in the EEG: elimination of false positives | |
| Wright et al. | Autoregression models of EEG: results compared with expectations for a multilinear near-equilibrium biophysical process | |
| Acharjee et al. | Independent vector analysis for gradient artifact removal in concurrent EEG-fMRI data | |
| CN113367657B (en) | Sleep quality evaluation method, device, equipment and storage medium based on high-frequency electroencephalogram | |
| Deng et al. | Complexity extraction of electroencephalograms in Alzheimer's disease with weighted-permutation entropy | |
| CN110840432A (en) | Multichannel electroencephalogram epilepsy automatic detection device based on one-dimensional CNN-LSTM | |
| Babadi et al. | DiBa: a data-driven Bayesian algorithm for sleep spindle detection | |
| CN102488516A (en) | Nonlinear electroencephalogram signal analysis method and device | |
| Chang et al. | Multivariate autoregressive models with exogenous inputs for intracerebral responses to direct electrical stimulation of the human brain | |
| AU2020103949A4 (en) | EEG Signal Mixed Noise Processing Method, Equipment and Storage Medium | |
| CN117064409B (en) | Method, device and terminal for evaluating transcranial direct current intervention stimulation effect in real time | |
| Vieira et al. | Understanding the design neurocognition of mechanical engineers when designing and problem-solving | |
| CN108478215A (en) | EEG Noise Cancellation, storage medium based on wavelet analysis and device | |
| Stevenson et al. | A nonlinear model of newborn EEG with nonstationary inputs | |
| Cerutti et al. | Analysis of visual evoked potentials through Wiener filtering applied to a small number of sweeps | |
| CN117349598B (en) | Electroencephalogram signal processing method and device, equipment and storage medium | |
| CN111820876A (en) | Dynamic construction method of electroencephalogram spatial filter | |
| CN112115856A (en) | Electroencephalogram quality evaluation method, storage medium and system | |
| CN114190953A (en) | Training method and system of electroencephalogram signal noise reduction model for electroencephalogram acquisition equipment | |
| CN117643474B (en) | Electroencephalogram signal processing method and system based on deep neural network | |
| Leach et al. | ‘High-Density-SleepCleaner’: An open-source, semi-automatic artifact removal routine tailored to high-density sleep EEG | |
| Demirel et al. | Single-channel EEG based arousal level estimation using multitaper spectrum estimation at low-power wearable devices | |
| CN115357126B (en) | Method, system and device for extracting sleep slow wave-spindle wave coupling signal | |
| Wei et al. | Automatic recognition of chewing noises in epileptic EEG based on period segmentation | |
| Asha et al. | Automated seizure detection from multichannel EEG signals using support vector machine and artificial neural networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |