CN117349598A - Electroencephalogram signal processing method and device, equipment and storage medium - Google Patents
Electroencephalogram signal processing method and device, equipment and storage medium Download PDFInfo
- Publication number
- CN117349598A CN117349598A CN202311643652.2A CN202311643652A CN117349598A CN 117349598 A CN117349598 A CN 117349598A CN 202311643652 A CN202311643652 A CN 202311643652A CN 117349598 A CN117349598 A CN 117349598A
- Authority
- CN
- China
- Prior art keywords
- signal
- index
- determining
- calculating
- signals
- 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.)
- Granted
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 63
- 238000007781 pre-processing Methods 0.000 claims abstract description 47
- 238000012958 reprocessing Methods 0.000 claims abstract description 6
- 238000005070 sampling Methods 0.000 claims description 119
- 210000004556 brain Anatomy 0.000 claims description 32
- 230000033764 rhythmic process Effects 0.000 claims description 26
- 238000000034 method Methods 0.000 claims description 24
- 238000009827 uniform distribution Methods 0.000 claims description 19
- 238000004364 calculation method Methods 0.000 claims description 16
- 230000008859 change Effects 0.000 claims description 15
- 238000009826 distribution Methods 0.000 claims description 15
- 238000011156 evaluation Methods 0.000 claims description 15
- 238000010183 spectrum analysis Methods 0.000 claims description 10
- 238000007499 fusion processing Methods 0.000 claims description 9
- 239000002131 composite material Substances 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 6
- 238000003491 array Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 15
- 238000000926 separation method Methods 0.000 abstract description 3
- 238000001914 filtration Methods 0.000 description 18
- 239000011159 matrix material Substances 0.000 description 9
- 230000008569 process Effects 0.000 description 8
- 238000001228 spectrum Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 6
- 238000000354 decomposition reaction Methods 0.000 description 6
- 210000003128 head Anatomy 0.000 description 6
- 238000012880 independent component analysis Methods 0.000 description 6
- 238000005457 optimization Methods 0.000 description 6
- 230000009466 transformation Effects 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 210000004761 scalp Anatomy 0.000 description 3
- 238000013097 stability assessment Methods 0.000 description 3
- 108010076504 Protein Sorting Signals Proteins 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 238000012806 monitoring device Methods 0.000 description 2
- 230000003183 myoelectrical effect Effects 0.000 description 2
- 238000011158 quantitative evaluation Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000000717 retained effect Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 208000014644 Brain disease Diseases 0.000 description 1
- 101100433290 Homo sapiens ZNF471 gene Proteins 0.000 description 1
- 108010001267 Protein Subunits Proteins 0.000 description 1
- 101100389697 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) ERP1 gene Proteins 0.000 description 1
- 101100333756 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) ERP2 gene Proteins 0.000 description 1
- 101100066419 Xenopus laevis fbxo43 gene Proteins 0.000 description 1
- 102100029037 Zinc finger protein 471 Human genes 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 230000006998 cognitive state Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000005281 excited state Effects 0.000 description 1
- 210000001652 frontal lobe Anatomy 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000007769 metal material Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 230000001936 parietal effect Effects 0.000 description 1
- 238000010827 pathological analysis Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000008667 sleep stage Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 210000003478 temporal lobe Anatomy 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- 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
- A61B5/374—Detecting the frequency distribution of signals, e.g. detecting delta, theta, alpha, beta or gamma waves
-
- 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
- A61B5/7221—Determining signal validity, reliability or quality
-
- 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
- A61B5/7253—Details of waveform analysis characterised by using transforms
- A61B5/7257—Details of waveform analysis characterised by using transforms using Fourier transforms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2431—Multiple classes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Heart & Thoracic Surgery (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Psychiatry (AREA)
- Surgery (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Medical Informatics (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Psychology (AREA)
- Physiology (AREA)
- Signal Processing (AREA)
- Mathematical Physics (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention belongs to the technical field of brain-computer interfaces, and discloses an electroencephalogram signal processing method, an electroencephalogram signal processing device, equipment and a storage medium, wherein preprocessing signals are obtained by collecting electroencephalogram physiological signals and preprocessing, a plurality of indexes of the preprocessing signals are calculated and fused to obtain comprehensive indexes, and corresponding signal stability indexes are determined; simultaneously calculating a signal strength index of the preprocessed signal; according to the signal stability index and the signal strength index, calculating a signal quality score, determining a corresponding quality grade, and finally reprocessing the preprocessed signal according to a processing strategy corresponding to the quality grade, so that the quality of the signal can be quantified, and through two-step separation processing, the information is reserved through general preprocessing, then the quality is adaptively improved according to the signal quality score, the loss of effective information caused by excessive processing can be avoided, more effective information is reserved, the information availability is improved, the processing effect is improved, and the signal quality is effectively improved.
Description
Technical Field
The invention belongs to the technical field of brain-computer interfaces, and particularly relates to an electroencephalogram signal processing method and device, equipment and a storage medium.
Background
The electroencephalogram signal is a signal for recording brain electrical activity, and the frequency range covers a plurality of frequency bands such as δ, θ, α, β, γ, and the like. The brain electrical signal can be used for clinical auxiliary diagnosis of various brain diseases, can reflect the cognitive state change of individuals, and is widely applied to scientific researches in the fields of brain-computer interfaces, nerve marketing and the like. However, the original electroencephalogram signals obtained by direct collection often mix with a large amount of noise, and the sources of the noise are complex, including physiological noise such as brain background activity, electrocardiographic trouble and electrooculogram trouble, and non-physiological noise such as drifting noise caused by electrode movement, environmental electromagnetic noise and the like. The signal-to-noise ratio of the original brain electrical signal is generally low, and the direct application of the original brain electrical signal to pathological analysis or individual state evaluation can have great difficulty.
The objective of electroencephalogram signal processing is to improve the usability of signals, and various noises need to be removed by adopting an effective method to improve the signal quality. The current processing method mainly comprises analog filtering, digital filtering, independent component analysis and the like. Most of the processes use a unified preset process flow, and the quality condition of the signal is not considered. The signal with better quality is excessively processed to cause effective information loss; insufficient signal processing with poor quality can not effectively improve quality. Therefore, the processing effect is not good enough, and the quality of the processed signal is still insufficient.
Disclosure of Invention
The invention aims to provide an electroencephalogram signal processing method, an electroencephalogram signal processing device, electroencephalogram signal processing equipment and a storage medium, which can improve the processing effect and effectively improve the signal quality.
The first aspect of the invention discloses an electroencephalogram signal processing method, which comprises the following steps:
collecting brain electric physiological signals;
preprocessing the acquired electroencephalogram physiological signals to obtain preprocessed signals;
calculating a plurality of indexes of the preprocessing signal, and carrying out fusion processing on the indexes to obtain a comprehensive index;
determining a corresponding signal stability index according to the comprehensive index;
calculating a signal strength index of the preprocessed signal;
calculating a signal quality score based on the signal stability index and the signal strength index;
determining a corresponding quality level according to the signal quality score;
and reprocessing the preprocessed signals according to a processing strategy corresponding to the quality level.
In some embodiments, the plurality of indices includes a frequency fluctuation index, a functional connection stability assessment index, and an event-related potential waveform change index; calculating a plurality of indexes of the preprocessing signal, and carrying out fusion processing on the plurality of indexes to obtain a comprehensive index, wherein the method comprises the following steps:
Detecting the frequency range of the preprocessing signal, and calculating a frequency fluctuation index;
calculating a functional connection stability evaluation index of the preprocessed signals;
calculating an event-related potential waveform change index of the preprocessing signal;
and carrying out normalization processing on the frequency fluctuation index, the functional connection stability evaluation index and the event-related potential waveform change index, and calculating to obtain the comprehensive index of the preprocessing signal.
In some embodiments, detecting the frequency range of the pre-processed signal, calculating a frequency fluctuation index, comprises:
performing fast Fourier transform on the preprocessing signal to obtain a spectrum analysis result;
detecting rhythm components in the electroencephalogram signals in the spectrum analysis result, and determining a frequency range corresponding to each rhythm component;
counting the maximum amplitude and the minimum amplitude in each frequency range, and determining a fluctuation value according to the difference value of the maximum amplitude and the minimum amplitude;
and counting the number duty ratio of the frequency range of which the fluctuation value exceeds a preset fluctuation threshold value, and determining a frequency fluctuation index according to the number duty ratio.
In some embodiments, determining a corresponding signal stability index from the composite index comprises:
If the integrated index is less than or equal to a first threshold value, determining that the signal stability index is 1;
if the integrated index is greater than the first threshold and less than or equal to a second threshold, determining a signal stability index of 2;
if the integrated index is greater than the second threshold and less than or equal to a third threshold, determining a signal stability index of 3;
and if the integrated index is greater than the third threshold value, determining that the signal stability index is 4.
In some embodiments, calculating a signal strength index of the pre-processed signal comprises:
controlling a sliding window with a specified length to slide and divide signals on the preprocessing signals one by one to obtain a plurality of time windows, and extracting rhythm component signals from the electroencephalogram signals in each time window;
calculating sampling amplitude values of sampling points of the rhythm component signals in each time window, and storing the sampling amplitude values into a window array corresponding to the time window;
combining window arrays of all time windows into a total sampling amplitude array, determining a sampling amplitude interval to which the sampling amplitude of each sampling point in the total sampling amplitude array belongs, and counting the number proportion of the sampling points in each sampling amplitude interval;
And determining the signal strength index according to the number of sampling points in all the sampling amplitude intervals.
In some embodiments, determining the signal strength index from the number of sampling points in the total sampling amplitude interval to the ratio comprises:
generating a corresponding target distribution array according to the sampling point quantity proportion in all the sampling amplitude intervals;
acquiring a preset uniform distribution array, wherein the uniform distribution array comprises expected duty ratios of a plurality of standard intervals under uniform distribution, the number of the plurality of standard intervals is the same as the number of all sampling amplitude intervals, and the plurality of standard intervals are obtained by uniformly dividing the whole sampling amplitude range;
and calculating the Euclidean distance between the target distribution array and the uniform distribution array, and determining a signal intensity index according to the Euclidean distance.
In some embodiments, determining a signal strength index from the euclidean distance comprises:
if the Euclidean distance is less than or equal to 0.05, the signal strength index is 1;
if the Euclidean distance is more than 0.05 and less than or equal to 0.1, the signal intensity index is 2;
if the Euclidean distance is more than 0.1 and less than or equal to 0.3, the signal intensity index is 3;
If the Euclidean distance is greater than 0.3 and less than or equal to 0.35, the signal strength index is 4
The second aspect of the present invention discloses an electroencephalogram signal processing apparatus, comprising:
the acquisition unit is used for acquiring brain electric physiological signals;
the first processing unit is used for preprocessing the acquired electroencephalogram physiological signals to obtain preprocessed signals;
the first calculating unit is used for calculating a plurality of indexes of the preprocessing signal and obtaining a comprehensive index through fusion processing of the indexes;
the first determining unit is used for determining a corresponding signal stability index according to the comprehensive index;
a second calculation unit for calculating a signal strength index of the preprocessed signal;
a third calculation unit for calculating a signal quality score according to the signal stability index and the signal strength index;
a second determining unit, configured to determine a corresponding quality level according to the signal quality score;
and the second processing unit is used for reprocessing the preprocessed signals according to a processing strategy corresponding to the quality grade.
A third aspect of the invention discloses an electronic device comprising a memory storing executable program code and a processor coupled to the memory; the processor invokes the executable program code stored in the memory for executing the electroencephalogram signal processing method disclosed in the first aspect.
A fourth aspect of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program causes a computer to execute the electroencephalogram signal processing method disclosed in the first aspect.
The method has the advantages that preprocessing signals are obtained by acquiring the electroencephalogram physiological signals for preprocessing, then a plurality of indexes of the preprocessing signals are calculated for fusion processing to obtain comprehensive indexes, and corresponding signal stability indexes are determined; simultaneously calculating a signal strength index of the preprocessed signal; and calculating a signal quality score according to the signal stability index and the signal strength index, determining a corresponding quality grade, and finally re-processing the preprocessed signal according to a processing strategy corresponding to the quality grade, so that the quality of the signal can be quantified through the signal quality score, and the signal is subjected to two-step separation processing, the information is retained through general preprocessing, then the quality is adaptively improved according to the signal quality score, so that the loss of effective information caused by excessive processing can be avoided, more effective information is retained, the information availability is improved, the processing effect is improved, the signal quality is effectively improved, and the analyzability of the processed signal is stronger.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles and effects of the invention.
Unless specifically stated or otherwise defined, the same reference numerals in different drawings denote the same or similar technical features, and different reference numerals may be used for the same or similar technical features.
Fig. 1 is a flowchart of an electroencephalogram signal processing method disclosed in an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an electroencephalogram signal processing apparatus according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Reference numerals illustrate:
201. an acquisition unit; 202. a first processing unit; 203. a first calculation unit; 204. a first determination unit; 205. a second calculation unit; 206. a third calculation unit; 207. a second determination unit; 208. a second processing unit; 301. a memory; 302. a processor.
Detailed Description
Unless defined otherwise or otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. In the context of a realistic scenario in connection with the technical solution of the invention, all technical and scientific terms used herein may also have meanings corresponding to the purpose of the technical solution of the invention. The terms "first and second …" are used herein merely for distinguishing between names and not for describing a particular number or order. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.
It will be understood that when an element is referred to as being "fixed" to another element, it can be directly fixed to the other element or intervening elements may also be present; when an element is referred to as being "connected" to another element, it can be directly connected to the other element or intervening elements may also be present; when an element is referred to as being "mounted to" another element, it can be directly mounted to the other element or intervening elements may also be present. When an element is referred to as being "disposed on" another element, it can be directly on the other element or intervening elements may also be present.
As used herein, unless specifically stated or otherwise defined, "the" means that the feature or technical content mentioned or described before in the corresponding position may be the same or similar to the feature or technical content mentioned. Furthermore, the terms "comprising," "including," and "having," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed steps or elements but may include other steps or elements not listed or inherent to such process, method, article, or apparatus.
The embodiment of the invention discloses an electroencephalogram signal processing method which can be realized through computer programming. The execution main body of the method can be electronic equipment such as a computer, a notebook computer, a tablet computer and the like, or an electroencephalogram signal processing device embedded in the electronic equipment, and the invention is not limited to the above. In order that the invention may be readily understood, a more particular description of specific embodiments thereof will be rendered by reference to specific embodiments that are illustrated in the appended drawings.
As shown in FIG. 1, the method comprises the following steps 110-170:
110. and collecting brain electric physiological signals.
The brain electric physiological signals of the user are collected through brain electric monitoring devices worn by the user, such as brain electric head rings or brain electric head sleeves. The electroencephalogram monitoring device is provided with a plurality of electrodes, the electrodes are made of metal materials and are connected with all channels of the acquisition system, and the distance between the electrodes is kept at 3cm according to the international 10-20 system arrangement. In order to obtain a high signal-to-noise ratio signal, the structural design of the head ring is matched with the natural curvature of the head, the electrode is firmly attached to the scalp by adjusting the binding band, and the skin contact effect of the electrode can be optimized by slightly rotating the head ring. In parameter setting, a 0.1-100Hz broadband filter is adopted to remove the environmental electromagnetic interference. The alpha wave is provided with 8-13Hz filter band, and the alpha wave appears in a relaxed eye-closing state, and the peak frequency is about 10Hz, which represents the slowing down of the brain electrical activity. The alpha wave is small in amplitude, about 20-60 mu V. The sampling frequency is not less than 30Hz to meet the Nyquist theorem. In addition, considering that the amplitude of the α -wave is small in the range of 20 to 60 μv, the gain of the amplifier needs to be properly increased, and the analog-to-digital conversion accuracy also needs to be 24 bits or more to obtain an α -wave signal of sufficient resolution. The beta wave represents an excited state, has a frequency of 13-30Hz and an amplitude of 2-20 mu V, and is concentrated in the frontal lobe area. The theta wave appears in the early stage of sleep, the frequency is 4-8Hz, and the amplitude is larger and can reach 50-100 mu V. The gamma wave is closely related to advanced cognition, the frequency is 30-100Hz, and the amplitude is about 5-10 mu V. The number of input channels of the acquisition system needs to be consistent with the number of electrodes arranged on the headband so that each electrode can acquire a signal as an independent channel.
120. Preprocessing the acquired electroencephalogram physiological signals to obtain preprocessed signals.
The method adopts a low-intensity general mode to carry out basic processing and mainly comprises the step of obtaining digital signals in a digital form by analog-to-digital conversion of analog acquired brain electrical signals. For compatibility between different devices, the digital signal needs to be converted into a standard file format. Meanwhile, whether the signal time sequence has the defect of a long time period is checked, a simple interpolation mode is used for supplementing missing sample points, and common interpolation modes are interpolation by using the average value of adjacent points.
In addition, due to sensor stability problems, there may be a dc drift in the signal sequence, which affects the comparability of the signals in different time periods, requiring correction with a simple high pass filter, eliminating baseline drift caused by drift. Then, the signal is subjected to low-pass filtering smoothing processing by adopting fixed parameters, so that high-frequency random noise caused by myoelectric interference can be reduced, but a filter with relaxed parameters is required to avoid possible information loss. Finally, the sampling frequency of the signal is appropriately reduced according to the requirements of analysis tasks, so that the storage and calculation amount is reduced. Through the low-intensity general preprocessing, the basic standardization processing of the signals is completed, and a foundation is laid for the subsequent optimization processing.
130. And calculating a plurality of indexes of the preprocessing signal, and carrying out fusion processing on the plurality of indexes to obtain a comprehensive index.
The plurality of indexes may include at least two combinations, preferably three combinations, of the frequency fluctuation index, the functional connection stability evaluation index, and the event-related potential waveform change index, and step 130 may include the following steps 1301 to 1304, which are not illustrated:
1301. the frequency range of the preprocessed signal is detected and a frequency fluctuation index is calculated.
Specifically, step 1301 includes the following steps S11 to S14, not shown:
s11, performing fast Fourier transform on the preprocessed signals to obtain a spectrum analysis result.
Specifically, parameters of the fast fourier transform (Fast Fourier Transformation, FFT) are first set, mainly the sampling frequency Fs and the number of samples N to be converted. The sampling frequency Fs needs to be set according to the nyquist sampling theorem, typically taking 256Hz or higher. The number of samples N takes the number of samples collected in the last 1s, and if fs=256 Hz, n=256. Then, the acquired preprocessed electroencephalogram signal is divided into N segments with the length, each segment is a sequence to be subjected to FFT, and the sequence is marked as x (N), and N is from 0 to N-1. For each sample sequence X (N) of length N, FFT is performed using an FFT algorithm to obtain its spectral representation X (k), k from 0 to N-1. The amplitude spectrum a (k) = |x (k) | of the spectrum X (k) continues to be calculated, and a (k) represents the energy distribution of the electroencephalogram signal at the corresponding frequency fk. According to the corresponding relation between the k value and the sampling frequency Fs An amplitude-frequency function diagram of the electroencephalogram signal in the sampling frequency range can be obtained. And repeating the process, performing FFT analysis on all the acquired electroencephalogram signal sequences, and finally obtaining the average energy spectrum of the full-segment signals as a spectrum analysis result.
S12, detecting rhythm components in the electroencephalogram signals in the spectrum analysis result, and determining a frequency range corresponding to each rhythm component.
The frequency range corresponding to the brain electrical rhythm can be analyzed and detected on the average energy spectrogram. For example, the alpha wave, beta wave, theta wave and the like which are main rhythm components in the electroencephalogram signals have responses in a specific frequency range. For example, alpha waves are typically concentrated in the 8-13Hz range, beta waves in the 13-30Hz range, and theta waves in the 4-8Hz range.
S13, counting the maximum amplitude and the minimum amplitude in each frequency range, and determining a fluctuation value according to the difference value of the maximum amplitude and the minimum amplitude.
By calculating the difference Δa=amax-Amin between the maximum amplitude Amax and the minimum amplitude Amin in each frequency range, Δa can represent the fluctuation in that frequency range. If ΔA is larger, it means that the more severe the fluctuation in this frequency range, the spectrum is not smooth enough. Therefore, ΔA for the frequency range of each rhythm component may be calculated and normalized to between 0 and 1 to obtain the fluctuation value ΔAi' for that frequency range.
S14, counting the number duty ratio of the frequency range of which the fluctuation value exceeds the preset fluctuation threshold value, and determining the frequency fluctuation index according to the number duty ratio.
The frequency fluctuation index FI may be directly a number of frequency ranges whose fluctuation value exceeds a preset fluctuation threshold. For example, if the number of defined frequency ranges is N, the fluctuation amplitude of the ith frequency range is Δai, normalization processing is performed to obtain Δai ', a preset fluctuation threshold T is set, and the number N of Δai' exceeding the threshold T is counted, then the frequency fluctuation index fi=n/N. If delta A is very small in all frequency ranges, the spectrum is stable; if the individual frequency range ΔA is too large, this range is interpreted as fluctuating abnormally, and the spectrum is not sufficiently stable. Therefore, if the frequency fluctuation index FI is larger, the spectrum is described as being less stable.
1302. And calculating a functional connection stability evaluation index of the pretreatment signal.
In order to evaluate the stability of the brain electrical signal in terms of functional connection, a mapping relationship between the scalp electrode and the brain region needs to be established, the electrode signals are mapped to the corresponding brain regions, and then the functional connection relationship between different brain regions is calculated. The specific method comprises the following steps: firstly, according to the electrode arrangement scheme of the international 10-20 system, the position of each electrode on the scalp is determined, and according to the relation between the distribution of brain areas and the projection of the electrodes, the mapping rule of the electrodes and the brain areas is set. For example, the occipital central brain region is associated with the O1, O2 electrodes, the temporal upper brain region is associated with the T3, T4 electrodes, etc. Then, the acquired preprocessing brain electrical signals are divided into a plurality of segments with the length of 10s, and the step length is set to be 5s, namely, the overlapping part of 5s exists between two adjacent segments of signals. For each segment of signal, calculating the linear correlation coefficient between every two electrode signals to form an electrode-electrode correlation matrix R. Then, according to the preset electrode-brain region mapping relation, the related matrix R is converted into a brain region-brain region functional connection matrix M. Each element in the functional connection matrix M represents the functional connection strength between two brain regions. By repeating this process, a series of time-adjacent functional connection matrices M1, M2, MT can be obtained. Where Mt represents the functional connection matrix at time T, t=1, 2, …, T. Then, a change coefficient Kt= |Mt-Mt-1|/|Mt| between each function connection matrix and the function connection matrix corresponding to the last time point adjacent to the function connection matrix is calculated. Where Mt represents the sum of the absolute values of all elements in the matrix Mt. Finally, the variation coefficients Kt of all the functional connection matrices are averaged to obtain an average variation coefficient MCV= (ΣKt)/(T-1), and the MCV reflects the average variation degree of the functional connection matrices in a period of time and can be used as an evaluation index of the functional connection stability.
1303. And calculating an event-related potential waveform change index of the preprocessing signal.
Event-related potential (ERP) is a repetitive waveform that appears in the brain electricity triggered by an external stimulus. To evaluate the stability of ERP, it is necessary to calculate the coefficient of variation of the ERP waveform over a period of time. The specific method comprises the following steps: firstly, extracting a trigger event, recording a trigger time point, and taking the time point as a 0 moment. Taking a fixed length time window, such as 1000ms, each trigger will obtain ERP waveform data of 1000ms in length.
Dividing the preprocessing signal into a plurality of sections of ERP waveforms according to the triggering event. Operating with the same electrode channel waveforms. Suppose N ERP waveforms { ERP1, ERP2, & ERPn }. For each time point t, calculating the mean value u (t) = (ERP 1 (t) +erp2 (t) +erpn (t))/N of the N ERP waveforms at that time point t; then, the standard deviation std (t) =sqrt (Σ (ERPn (t) -u (t))Σ2/(N-1)) is calculated, and the coefficient of variation at this time point t is cv (t) =std (t)/u (t).
The above equation is repeated to calculate the coefficient of variation cv (t) at each time point, and a coefficient of variation curve with the total length of 1000ms is obtained. Finally, calculating the average value of each point on the variation coefficient curve cv (t) as the variation coefficient of the channel ERP waveform: cv=mean (CV (t)).
The above steps were repeated to obtain ERP variation coefficients for each channel, noted { CV1, CV2,.. CVn }. Channels located in the critical brain region are selected from all channels as critical channels to emphasize ERP response changes in the critical brain region. For example: c3 and C4 are used for detecting the sports cortex area; p3 and P4 are used for detecting the parietal cortex area; t5, T6 are used to detect temporal lobe areas. And averaging the ERP variation coefficients of the key channels to obtain the average ERP variation coefficient of the key channels, and taking the average ERP variation coefficient of the key channels as a final ERP waveform variation index EKCV.
1304. And carrying out normalization processing on the frequency fluctuation index, the functional connection stability evaluation index and the event-related potential waveform change index, and calculating to obtain the comprehensive index of the preprocessing signal.
Optionally, the normalization formula of the plurality of indexes is shown in the following formula (1):
（1）
wherein SI is a composite index, FI, MCV, EKCV is a frequency fluctuation index, a functional connection stability evaluation index, and an event-related potential waveform change index, w, respectively 1 、w 2 、w 3 Each of which is a weight coefficient of FI, MCV, EKCV. The smaller the SI, the smaller the signal change, the better the signal stability.
Steps 1301-1304 are implemented, and through four aspects of quantitative evaluation of frequency stability, functional connection stability, ERP stability and intensity distribution uniformity, the reliability of evaluation indexes can be improved, and the signal quality can be reflected more accurately.
140. And determining a corresponding signal stability index according to the comprehensive index.
And comparing the comprehensive index SI with a preset threshold value to generate stability assessment values of different grades, namely a signal stability index RS. Specifically, the 3 thresholds are respectively a first threshold, a second threshold and a third threshold, wherein the first threshold is smaller than the second threshold, and the second threshold is smaller than the third threshold. And comparing the comprehensive index SI with a threshold value to obtain a signal stability index RS. If the integrated index SI is less than or equal to the first threshold, determining a signal stability index rs=1, indicating that the signal stability is excellent; if the integrated index SI is greater than the first threshold and less than or equal to the second threshold, determining a signal stability index rs=2, indicating that the signal stability is good; if the integrated index SI is greater than the second threshold and less than or equal to the third threshold, determining a signal stability index rs=3, indicating that the signal stability is general; if the integrated index SI is greater than the third threshold, a signal stability index rs=4 is determined, indicating poor signal stability.
For example, the first threshold is 0.2, the second threshold is 0.5, and the third threshold is 0.8. If SI < = 0.2, rs=1 (signal stability is excellent); if 0.2< si < = 0.5, rs=2 (signal stability is good); if 0.5< si < = 0.8, rs=3 (signal stability is general); if SI >0.8, rs=4 (signal stability is poor).
150. A signal strength index of the pre-processed signal is calculated.
1501. And controlling the sliding window with the designated length to slide and divide the signals on the preprocessing signals one by one to obtain a plurality of time windows, and extracting rhythm component signals from the electroencephalogram signals in each time window.
The window length of 2-10s is generally set as the appointed length, and the signals are divided by sliding on the preprocessing signals one by one according to the window length, so as to obtain a plurality of time windows. And extracting the rhythm component of the specific frequency band from the electroencephalogram signals in each divided time window by using a band-pass filter to obtain a rhythm component signal.
The rhythm components of interest include delta wave, theta wave, alpha wave, beta wave, etc., and the corresponding specific frequency bands are selected according to the rhythm components for filtering extraction, for example, the specific frequency band of delta wave is 0.5-4Hz, the specific frequency band of theta wave is 4-8Hz, the specific frequency band of alpha wave is 8-13Hz, the specific frequency band of beta wave is 13-30Hz, etc. In this way, the signals are divided through the sliding window and extracted through band-pass filtering, and the signals in the time window after filtering extraction contain the rhythm components of the specific frequency band, namely the obtained rhythm component signals can be used as input of subsequent analysis.
1502. And calculating the sampling amplitude of each sampling point of the rhythm component signal in each time window, and storing the sampling amplitude in a window array corresponding to the time window.
Where the sample amplitude ISM represents the signal strength magnitude at that point. Specifically, a plurality of sampling points of a rhythm component signal in a window are regarded as a sampling point sequence s (n), each sampling point is traversed according to the sequence order, a sampling value of each sampling point s (n) is extracted, and a sampling amplitude ISM is calculated. The calculation formula is ISM= |s (n) |, and the absolute value of the sampling value is taken as the sampling amplitude. And repeating the operations in sequence to obtain sampling amplitude ISM of all sampling points of the window. The sampling amplitude values are sequentially stored into a corresponding window array A_window to serve as a sampling amplitude value sequence of the window. Repeating the above processing until the computation and storage of the sampling amplitude ISM of all sampling points in all time windows are completed.
Assume that an electroencephalogram signal is recorded, the sampling frequency is 1000Hz, and 1000 sampling points are acquired in total. For analysis, the 1000 points are divided into 10 time windows, each window containing 100 sample points, and the sample amplitude of each sample point is calculated one by one: the 1 st sampling point has a value of 0.532 mV, the sampling amplitude of this point is |0.532|= 0.5322, the 2 nd sampling point has a value of-2.158 mV, the sampling amplitude of this point is | 2.158 |= 2.158, and so on, to obtain the sampling amplitudes of all 100 sampling points in the window, and finally the 100 sampling amplitudes are stored in the corresponding window array a_window.
1503. And merging the window arrays A_windows of all the time windows into a total sampling amplitude array AMP, determining a sampling amplitude interval to which the sampling amplitude of each sampling point in the total sampling amplitude array AMP belongs, and counting the number proportion of the sampling points in each sampling amplitude interval.
Specifically, the window arrays a_windows may be arranged and combined in succession into one large array AMP, i.e., amp= [ a_windows 1, a_windows 2, a_windows ]. After the AMP array storing all window sampling amplitudes is obtained, the array is traversed, and the sampling amplitude of each sampling point is sequentially extracted, and it is determined which sampling amplitude interval defined in advance, for example, 0-0.5, 0.5-1, etc. And counts num_rang1, num_rang2, etc. of the number of sampling points in each sampling amplitude interval. After traversing the AMP array, the total number of sampling points num_total of the array AMP is calculated. And finally, calculating the proportion p of the number of sampling points in each sampling amplitude interval to the total number, for example, p1=num_range 1/num_total, and repeatedly calculating to obtain the number of sampling points in all the sampling amplitude intervals to be p1, p 2.
1504. And determining the signal strength index according to the number of sampling points in all the sampling amplitude intervals.
Optionally, step 1504 may include the following steps S21 to S23, not shown:
s21, generating a corresponding target distribution array according to the sampling point quantity proportion in all the sampling amplitude intervals.
And generating a target distribution array P= [ P1, P2, ], pm ] for describing the sampling point number ratio in each sampling amplitude interval according to the sampling point number ratio P1, P2, & gt, pm of all the sampling amplitude intervals, wherein pi is the ratio of the sampling point number in the ith sampling amplitude interval to the total sampling point number.
S22, acquiring a preset uniform distribution array, wherein the uniform distribution array comprises expected duty ratios of a plurality of standard intervals under uniform distribution, the number of the standard intervals is the same as that of all sampling amplitude intervals, and the standard intervals are obtained by uniformly dividing the whole sampling amplitude range.
In the embodiment of the present invention, an evenly distributed array q= [ Q1, Q2, ], qm ] may be defined, where m represents the total number of standard intervals, and represents that the whole sampling amplitude range is evenly divided into m standard intervals; qi represents the desired duty cycle of the ith standard interval in a uniform distribution. Because of the uniform distribution, the ratio of each standard interval to the whole range should be equal, so the desired ratio of each standard interval is 1/total number of intervals m, so for the i-th standard interval, the ratio qi=1/m.
S23, calculating Euclidean distance between the target distribution array and the uniform distribution array, and determining a signal intensity index according to the Euclidean distance.
The euclidean distance of the target distribution array p= [ P1, P2,..pm ] and the uniform distribution array Q of the sampling amplitude interval duty ratio is calculated by the following formula (2):
d=sqrt(sum((pi-qi)^2))（2）
the signal strength index RP is then determined from the d value. Specifically, quantization criteria may be set, rp=1 if d < =0.05; rp=2 if 0.05< d < = 0.1; rp=3 if 0.1< d < =0.3; rp=4 if 0.3< d < =0.35. The smaller the RP, the more uniform the intensity distribution, and the better the signal quality. The larger the RP, the worse the signal quality, which indicates the presence of noise interference with non-uniform sampling strength.
160. And calculating a signal quality score according to the signal stability index and the signal strength index, and determining a corresponding quality grade according to the signal quality score.
Where the signal quality score q= (rs+rp)/2. The smaller the signal quality score, the higher the corresponding quality level. For example, in the embodiment of the present invention, the quality class division criteria are set as follows:
if Q is a score of 1-1.5, the signal quality is excellent, the signal can be marked as a first-level signal, the uniformity and stability indexes of the electroencephalogram signal are ideal, and abundant effective information can be obtained.
If Q is a score of 1.5-2.5, which represents good signal quality, it can be marked as a second level signal, with better uniformity and stability, useful information can be obtained, but there may be slight noise interference.
If Q is a 2.5-3 score, meaning that the signal quality is generally marked as a third level signal, there is some degree of non-uniformity and instability, and care is taken to the denoising process.
If Q is 3-4, the representative signal quality is poor, the signal can be marked as a fourth-level signal, the uniformity and the stability are weak, the noise interference is serious, and the available information is reduced.
If Q is a score of 4-5, which corresponds to poor signal quality, it may be marked as a fifth level signal, with severe non-uniformity and instability characteristics, and difficulty in obtaining effective information.
170. And reprocessing the preprocessed signals according to the processing strategies corresponding to the quality grades.
The quality grades comprise a first grade, a second grade, a third grade, a fourth grade or a fifth grade with the quality decreasing in sequence, and the processing strategies corresponding to the quality grades are different. It should be noted that, step 120 uses low-intensity general preprocessing to primarily normalize the signal and retain more original information. In step 170, different processing strategies are adopted for further processing, and customization optimization is performed on the basis of knowing the signal quality, so that targeted processing can be performed according to the signal state, and better denoising and quality improvement effects are achieved. Compared with the one-time processing, the method has the advantages that parameters can be comprehensively adjusted, the optimization difficulty is high, the two-step processing is adopted, the adjustment, the control and the optimization are easier, the algorithm parameters of each step can be adjusted according to the effect through the step-by-step processing, and the system optimization is simpler. Specifically, in step 170:
For excellent signals with quality scores of 1-1.5, the quality level is the first level, the low-intensity normalized pre-processing of step 120 above is sufficient, and no complex signal reconstruction or enhancement is required, and thus no further processing is performed.
For good signals with quality scores of 1.5-2.5, the quality grade is the first grade, and a strategy of combining customized filtering and adaptive filtering can be performed to improve the analyzability of the signals, specifically, the frequency range of main noise components in the signals is firstly identified through spectrum analysis, for example, myoelectric noise is mainly distributed at 0.1-10Hz. Then, a digital filter is designed in a targeted mode, noise components with known frequency are suppressed by adopting a band-stop mode, and a band-stop filter with the frequency of 10-20Hz is designed to suppress low-frequency noise. And extracting time-frequency domain characteristics of the signals, and designing an adaptive filter according to the time-frequency domain characteristics, wherein filtering parameters such as cut-off frequency can be adjusted according to the real-time statistical characteristics of the signals, so as to realize adaptive filtering. In the filtering process, parameters, such as the bandwidth of a filter interval, are set to be 2Hz, so that characteristic loss caused by excessive smoothing is prevented, and the moderate filtering strength is ensured. And simultaneously, comparing the frequency spectrum changes before and after filtering, evaluating the filtering effect, and performing parameter optimization. The strategy combining the customized filtering and the self-adaptive filtering can effectively inhibit specific noise and simultaneously keep effective information to the maximum extent, thereby further improving the analyzable value of the signal.
For a general signal with a quality score of 2.5-3, the quality level is the first level, and a denoising strategy combining wavelet transformation and ICA can be adopted to improve the signal analyzability. Specifically, the wavelet transformation is performed first, the Db4 wavelet basis is selected as the analysis wavelet transformation, 5-layer wavelet decomposition is performed, the low-frequency component containing useful information is extracted, and the wavelet decomposition can effectively eliminate high-frequency noise. And then ICA noise reduction is further carried out on the signals after wavelet denoising, the iteration number is set to be 200, and the convergence threshold is set to be 0.0001 by adopting a fast ICA algorithm. ICA can further separate signal sources, removing noise interference due to independent sources. Meanwhile, in algorithm implementation, parameters are controlled to be set, and loss of useful information caused by excessive denoising is prevented. The strategy of combining wavelet transformation with ICA algorithm can remarkably inhibit noise interference in signals and improve the extractable degree of signal components containing effective characteristics, thereby enhancing the signal availability value in subsequent analysis.
For the poor signal with the quality score of 3-4, the quality grade is the first grade, and the processing such as filtering, signal reconstruction and the like with larger intensity is needed, but the information quantity loss is larger. The specific strategies are as follows: the multi-layer decomposition is carried out by adopting wavelet transformation, a Db4 wavelet basis is selected for 8-layer wavelet decomposition, and as the main energy of the effective electroencephalogram signal is concentrated at low frequency and the high-frequency part is mainly noise, the multi-layer decomposition can obviously inhibit high-frequency noise; EEMD mode decomposition is carried out on the extracted low-frequency signals, the mode number is set to be 10, the standard deviation of white noise is added to be 0.2, and EEMD can separate different oscillation modes, so that effective information can be further extracted; on the basis of EEMD reconstructed signals, an FIR low-pass filter is designed for noise reduction, the cut-off frequency is set to be 40Hz, the order is 60, the FIR filtering can smooth signals, and residual noise is restrained; and finally, carrying out wavelet reconstruction on the filtered signals to obtain reconstructed and enhanced brain electrical signals. The strategy integrates various technologies to perform layer-by-layer treatment, can obviously improve the signal to noise ratio, but can also cause effective information loss to a certain extent.
For signals with poor quality of 4-5 scores, the quality grade is the first grade, the currently acquired signals can be directly abandoned, the enhancement processing effect is limited, and a large amount of effective information can be lost. At this time, the currently acquired data is completely abandoned, the state of the electroencephalogram head ring device can be checked, the contact of the sensor is adjusted, the attention is focused, the optimal electroencephalogram signal acquisition state is ensured, and then the data is acquired again.
In summary, by implementing the embodiment of the invention, the quality of the signal can be quantified through signal quality scoring, and the signal quality can be effectively improved through two-step separation processing, the information is reserved through general preprocessing, and then the quality is adaptively improved according to the signal quality scoring, so that the loss of effective information caused by excessive processing can be avoided, more effective information is reserved, the information availability is improved, the processing effect is improved, the signal quality is effectively improved, and the analyzability of the processed signal is higher. In addition, through four aspects of quantitative evaluation of frequency stability, functional connection stability, ERP stability and intensity distribution uniformity, the reliability of evaluation indexes can be improved, and the signal quality can be reflected more accurately.
As shown in fig. 2, an embodiment of the present invention discloses an electroencephalogram signal processing apparatus, which includes an acquisition unit 201, a first processing unit 202, a first calculation unit 203, a first determination unit 204, a second calculation unit 205, a third calculation unit 206, a second determination unit 207, and a second processing unit 208, wherein,
an acquisition unit 201, configured to acquire an electroencephalogram physiological signal;
a first processing unit 202, configured to perform preprocessing on the acquired electroencephalogram signals to obtain preprocessed signals;
a first calculating unit 203, configured to calculate a plurality of indexes of the preprocessed signal, and perform fusion processing on the plurality of indexes to obtain a comprehensive index;
a first determining unit 204, configured to determine a corresponding signal stability index according to the composite index;
a second calculation unit 205 for calculating a signal strength index of the preprocessed signal;
a third calculation unit 206, configured to calculate a signal quality score according to the signal stability index and the signal strength index;
a second determining unit 207, configured to determine a corresponding quality level according to the signal quality score;
the second processing unit 208 is configured to reprocess the preprocessed signal according to a processing strategy corresponding to the quality level.
As an alternative embodiment, the plurality of indices includes a frequency fluctuation index, a functional connection stability assessment index, and an event-related potential waveform change index; the first computing unit 203 includes the following sub-units not shown:
The detection subunit is used for detecting the frequency range of the preprocessing signal and calculating a frequency fluctuation index;
a first calculation subunit for calculating a functional connection stability evaluation index of the preprocessed signal;
a second calculation subunit for calculating an event-related potential waveform change index of the preprocessing signal;
and the fusion subunit is used for carrying out normalization processing on the frequency fluctuation index, the functional connection stability evaluation index and the event-related potential waveform change index, and calculating to obtain the comprehensive index of the preprocessing signal.
Further optionally, the detection subunit is specifically configured to perform fast fourier transform on the preprocessed signal to obtain a spectrum analysis result; detecting rhythm components in the electroencephalogram signal in a frequency spectrum analysis result, and determining a frequency range corresponding to each rhythm component; counting the maximum amplitude and the minimum amplitude in each frequency range, and determining a fluctuation value according to the difference value of the maximum amplitude and the minimum amplitude; and counting the number duty ratio of the frequency range of which the fluctuation value exceeds the preset fluctuation threshold value, and determining the frequency fluctuation index according to the number duty ratio.
As an alternative embodiment, the first determining unit 204 is specifically configured to determine that the signal stability index is 1 when the composite index is less than or equal to the first threshold; and determining that the signal stability index is 2 when the composite index is greater than the first threshold and less than or equal to the second threshold; and determining that the signal stability index is 3 when the composite index is greater than the second threshold and less than or equal to the third threshold; and determining that the signal stability index is 4 when the composite index is greater than the third threshold.
As an alternative embodiment, the second computing unit 205 comprises the following sub-units, not shown:
the segmentation subunit is used for controlling the sliding window with the specified length to slide the segmentation signals one by one on the preprocessing signals so as to obtain a plurality of time windows;
an extraction subunit, configured to extract a rhythm component signal from the electroencephalogram signal in each time window;
the storage subunit is used for calculating the sampling amplitude of each sampling point of the rhythm component signal in each time window and storing the sampling amplitude into a window array corresponding to the time window;
the statistics subunit is used for combining window combinations of all time windows into a total sampling amplitude array, determining a sampling amplitude interval to which the sampling amplitude of each sampling point in the total sampling amplitude array belongs, and counting the number proportion of the sampling points in each sampling amplitude interval;
and the determining subunit is used for determining the signal strength index according to the number of the sampling points in all the sampling amplitude intervals.
Further optionally, the determining subunit is specifically configured to generate a corresponding target distribution array according to the number of sampling points in all the sampling amplitude intervals; the method comprises the steps of obtaining a preset uniform distribution array, wherein the uniform distribution array comprises expected duty ratios of a plurality of standard intervals under uniform distribution, the number of the standard intervals is the same as that of all sampling amplitude intervals, and the standard intervals are obtained by uniformly dividing the whole sampling amplitude range; and finally, calculating the Euclidean distance between the target distribution array and the uniform distribution array, and determining the signal intensity index according to the Euclidean distance.
Further optionally, the determining subunit is configured to determine the signal strength index according to the euclidean distance, specifically, if the euclidean distance is less than or equal to 0.05, the signal strength index is 1; if the Euclidean distance is more than 0.05 and less than or equal to 0.1, the signal intensity index is 2; if the Euclidean distance is more than 0.1 and less than or equal to 0.3, the signal intensity index is 3; if the Euclidean distance is greater than 0.3 and less than or equal to 0.35, the signal strength index is 4.
As shown in fig. 3, an embodiment of the present invention discloses an electronic device comprising a memory 301 storing executable program code and a processor 302 coupled to the memory 301;
the processor 302 calls executable program codes stored in the memory 301, and executes the electroencephalogram signal processing method described in the above embodiments.
The embodiments of the present invention also disclose a computer-readable storage medium storing a computer program, wherein the computer program causes a computer to execute the electroencephalogram signal processing method described in each of the above embodiments.
The foregoing embodiments are provided for the purpose of exemplary reproduction and deduction of the technical solution of the present invention, and are used for fully describing the technical solution, the purpose and the effects of the present invention, and are used for enabling the public to understand the disclosure of the present invention more thoroughly and comprehensively, and are not used for limiting the protection scope of the present invention.
The above examples are also not an exhaustive list based on the invention, and there may be a number of other embodiments not listed. Any substitutions and modifications made without departing from the spirit of the invention are within the scope of the invention.
Claims (10)
1. The electroencephalogram signal processing method is characterized by comprising the following steps of:
collecting brain electric physiological signals;
preprocessing the acquired electroencephalogram physiological signals to obtain preprocessed signals;
calculating a plurality of indexes of the preprocessing signal, and carrying out fusion processing on the indexes to obtain a comprehensive index;
determining a corresponding signal stability index according to the comprehensive index;
calculating a signal strength index of the preprocessed signal;
calculating a signal quality score based on the signal stability index and the signal strength index;
determining a corresponding quality level according to the signal quality score;
and reprocessing the preprocessed signals according to a processing strategy corresponding to the quality level.
2. The electroencephalogram signal processing method according to claim 1, wherein the plurality of indices includes a frequency fluctuation index, a functional connection stability evaluation index, and an event-related potential waveform variation index; calculating a plurality of indexes of the preprocessing signal, and carrying out fusion processing on the plurality of indexes to obtain a comprehensive index, wherein the method comprises the following steps:
Detecting the frequency range of the preprocessing signal, and calculating a frequency fluctuation index;
calculating a functional connection stability evaluation index of the preprocessed signals;
calculating an event-related potential waveform change index of the preprocessing signal;
and carrying out normalization processing on the frequency fluctuation index, the functional connection stability evaluation index and the event-related potential waveform change index, and calculating to obtain the comprehensive index of the preprocessing signal.
3. The electroencephalogram signal processing method according to claim 2, wherein detecting the frequency range of the preprocessing signal, calculating a frequency fluctuation index, comprises:
performing fast Fourier transform on the preprocessing signal to obtain a spectrum analysis result;
detecting rhythm components in the electroencephalogram signals in the spectrum analysis result, and determining a frequency range corresponding to each rhythm component;
counting the maximum amplitude and the minimum amplitude in each frequency range, and determining a fluctuation value according to the difference value of the maximum amplitude and the minimum amplitude;
and counting the number duty ratio of the frequency range of which the fluctuation value exceeds a preset fluctuation threshold value, and determining a frequency fluctuation index according to the number duty ratio.
4. The method of processing an electroencephalogram signal according to any one of claims 1 to 3, wherein determining a corresponding signal stability index from the composite index comprises:
If the integrated index is less than or equal to a first threshold value, determining that the signal stability index is 1;
if the integrated index is greater than the first threshold and less than or equal to a second threshold, determining a signal stability index of 2;
if the integrated index is greater than the second threshold and less than or equal to a third threshold, determining a signal stability index of 3;
and if the integrated index is greater than the third threshold value, determining that the signal stability index is 4.
5. The electroencephalogram signal processing method according to any one of claims 1 to 3, characterized in that calculating a signal intensity index of the preprocessing signal includes:
controlling a sliding window with a specified length to slide and divide signals on the preprocessing signals one by one to obtain a plurality of time windows, and extracting rhythm component signals from the electroencephalogram signals in each time window;
calculating sampling amplitude values of sampling points of the rhythm component signals in each time window, and storing the sampling amplitude values into a window array corresponding to the time window;
combining window arrays of all time windows into a total sampling amplitude array, determining a sampling amplitude interval to which the sampling amplitude of each sampling point in the total sampling amplitude array belongs, and counting the number proportion of the sampling points in each sampling amplitude interval;
And determining the signal strength index according to the number of sampling points in all the sampling amplitude intervals.
6. The method of electroencephalogram signal processing according to claim 5, wherein determining the signal strength index from the number of sampling points in the total sampling amplitude interval to the ratio comprises:
generating a corresponding target distribution array according to the sampling point quantity proportion in all the sampling amplitude intervals;
acquiring a preset uniform distribution array, wherein the uniform distribution array comprises expected duty ratios of a plurality of standard intervals under uniform distribution, the number of the plurality of standard intervals is the same as the number of all sampling amplitude intervals, and the plurality of standard intervals are obtained by uniformly dividing the whole sampling amplitude range;
and calculating the Euclidean distance between the target distribution array and the uniform distribution array, and determining a signal intensity index according to the Euclidean distance.
7. The method of electroencephalogram signal processing according to claim 6, wherein determining a signal strength index from the euclidean distance comprises:
if the Euclidean distance is less than or equal to 0.05, the signal strength index is 1;
if the Euclidean distance is more than 0.05 and less than or equal to 0.1, the signal intensity index is 2;
If the Euclidean distance is more than 0.1 and less than or equal to 0.3, the signal intensity index is 3;
if the Euclidean distance is greater than 0.3 and less than or equal to 0.35, the signal strength index is 4.
8. An electroencephalogram signal processing apparatus, comprising:
the acquisition unit is used for acquiring brain electric physiological signals;
the first processing unit is used for preprocessing the acquired electroencephalogram physiological signals to obtain preprocessed signals;
the first calculating unit is used for calculating a plurality of indexes of the preprocessing signal and obtaining a comprehensive index through fusion processing of the indexes;
the first determining unit is used for determining a corresponding signal stability index according to the comprehensive index;
a second calculation unit for calculating a signal strength index of the preprocessed signal;
a third calculation unit for calculating a signal quality score according to the signal stability index and the signal strength index;
a second determining unit, configured to determine a corresponding quality level according to the signal quality score;
and the second processing unit is used for reprocessing the preprocessed signals according to a processing strategy corresponding to the quality grade.
9. An electronic device comprising a memory storing executable program code and a processor coupled to the memory; the processor invokes the executable program code stored in the memory for performing the electroencephalogram signal processing method of any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, wherein the computer program causes a computer to execute the electroencephalogram signal processing method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311643652.2A CN117349598B (en) | 2023-12-04 | 2023-12-04 | Electroencephalogram signal processing method and device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311643652.2A CN117349598B (en) | 2023-12-04 | 2023-12-04 | Electroencephalogram signal processing method and device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117349598A true CN117349598A (en) | 2024-01-05 |
| CN117349598B CN117349598B (en) | 2024-03-08 |
Family
ID=89371432
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311643652.2A Active CN117349598B (en) | 2023-12-04 | 2023-12-04 | Electroencephalogram signal processing method and device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117349598B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117633442A (en) * | 2024-01-24 | 2024-03-01 | 小舟科技有限公司 | Brain-computer interface signal preprocessing method and computer equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004034885A2 (en) * | 2002-10-15 | 2004-04-29 | Medtronic Inc. | Signal quality monitoring and control for a medical device system |
| CN107273924A (en) * | 2017-06-06 | 2017-10-20 | 上海电力学院 | The Fault Analysis of Power Plants method of multi-data fusion based on fuzzy cluster analysis |
| CN109195178A (en) * | 2018-11-19 | 2019-01-11 | 国网河北省电力有限公司电力科学研究院 | signal quality monitoring method and terminal device |
| CN113288175A (en) * | 2021-05-31 | 2021-08-24 | 量心科技(深圳)有限公司 | Electroencephalogram signal quality detection method and device, electronic equipment and storage medium |
| WO2021180028A1 (en) * | 2020-03-10 | 2021-09-16 | 中国科学院脑科学与智能技术卓越创新中心 | Method, apparatus and device for evaluating sleep quality on basis of high-frequency electroencephalography, and storage medium |
| CN114795248A (en) * | 2022-04-15 | 2022-07-29 | 北京脑陆科技有限公司 | Electroencephalogram signal quality evaluation method and device, electronic equipment and medium |
| CN116500539A (en) * | 2023-04-25 | 2023-07-28 | 北京云恒科技研究院有限公司 | Evaluation and verification method for quality of TACAN ground beacon signal |
-
2023
- 2023-12-04 CN CN202311643652.2A patent/CN117349598B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004034885A2 (en) * | 2002-10-15 | 2004-04-29 | Medtronic Inc. | Signal quality monitoring and control for a medical device system |
| CN107273924A (en) * | 2017-06-06 | 2017-10-20 | 上海电力学院 | The Fault Analysis of Power Plants method of multi-data fusion based on fuzzy cluster analysis |
| CN109195178A (en) * | 2018-11-19 | 2019-01-11 | 国网河北省电力有限公司电力科学研究院 | signal quality monitoring method and terminal device |
| WO2021180028A1 (en) * | 2020-03-10 | 2021-09-16 | 中国科学院脑科学与智能技术卓越创新中心 | Method, apparatus and device for evaluating sleep quality on basis of high-frequency electroencephalography, and storage medium |
| CN113288175A (en) * | 2021-05-31 | 2021-08-24 | 量心科技(深圳)有限公司 | Electroencephalogram signal quality detection method and device, electronic equipment and storage medium |
| CN114795248A (en) * | 2022-04-15 | 2022-07-29 | 北京脑陆科技有限公司 | Electroencephalogram signal quality evaluation method and device, electronic equipment and medium |
| CN116500539A (en) * | 2023-04-25 | 2023-07-28 | 北京云恒科技研究院有限公司 | Evaluation and verification method for quality of TACAN ground beacon signal |
Non-Patent Citations (1)
| Title |
|---|
| 梁作玉: "脑机接口信号处理方法及控制应用研究", 《中国优秀硕士学位论文全文数据库》, no. 06, pages 1 - 84 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117633442A (en) * | 2024-01-24 | 2024-03-01 | 小舟科技有限公司 | Brain-computer interface signal preprocessing method and computer equipment |
| CN117633442B (en) * | 2024-01-24 | 2024-05-10 | 小舟科技有限公司 | Brain-computer interface signal preprocessing method and computer equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117349598B (en) | 2024-03-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2019100565A1 (en) | Method and device for self-learning dynamic electrocardiography analysis employing artificial intelligence | |
| US7672717B1 (en) | Method and system for the denoising of large-amplitude artifacts in electrograms using time-frequency transforms | |
| CN107174262B (en) | Attention evaluation method and system | |
| CA2849294C (en) | Physiological signal denoising | |
| CN105796096B (en) | A kind of heart rate variance analyzing method, system and terminal | |
| CN117349598B (en) | Electroencephalogram signal processing method and device, equipment and storage medium | |
| CA2584503A1 (en) | Monitoring physiological activity using partial state space reconstruction | |
| CN103610461A (en) | EEG noise elimination method based on dual-density wavelet neighborhood related threshold processing | |
| Patro et al. | De-noising of ECG raw signal by cascaded window based digital filters configuration | |
| Kumar et al. | Optimal selection of wavelet function and decomposition level for removal of ECG signal artifacts | |
| Tamburro et al. | Automatic removal of cardiac interference (ARCI): A new approach for EEG data | |
| Krusienski et al. | BCI signal processing: feature extraction | |
| Bortolan et al. | Dynamic filtration of high-frequency noise in ECG signal | |
| CN112294340A (en) | Method, system, storage medium and computer equipment for quickly and automatically removing myoelectric artifacts | |
| Dora et al. | Efficient detection and correction of variable strength ECG artifact from single channel EEG | |
| Islam et al. | Probability mapping based artifact detection and wavelet denoising based artifact removal from scalp EEG for BCI applications | |
| Jenkal et al. | Enhanced algorithm for QRS detection using discrete wavelet transform (DWT) | |
| CN113694374A (en) | Prediction system, method, device, processor and storage medium for predicting efficacy of treatment of obsessive-compulsive disorder based on EEG (electroencephalogram) | |
| Hegde et al. | A review on ECG signal processing and HRV analysis | |
| Jokić et al. | An efficient ECG modeling for heartbeat classification | |
| Dembrani et al. | Accurate detection of ECG signals in ECG monitoring systems by eliminating the motion artifacts and improving the signal quality using SSG filter with DBE | |
| Kiamini et al. | A wavelet based algorithm for ocular artifact detection in the EEG signals | |
| Rampil | Elements of EEG signal processing | |
| Arpaia et al. | Low-density EEG correction with multivariate decomposition and subspace reconstruction | |
| Gaurav et al. | Entropy based EEG irregularity quantification in single-channel SSVEP-based BCIS |
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 |