CN113791691B - Electroencephalogram signal band positioning method and device - Google Patents
Electroencephalogram signal band positioning method and device Download PDFInfo
- Publication number
- CN113791691B CN113791691B CN202111111846.9A CN202111111846A CN113791691B CN 113791691 B CN113791691 B CN 113791691B CN 202111111846 A CN202111111846 A CN 202111111846A CN 113791691 B CN113791691 B CN 113791691B
- Authority
- CN
- China
- Prior art keywords
- confidence
- electroencephalogram signal
- positioning
- confidence coefficient
- data
- 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
- 238000000034 method Methods 0.000 title claims abstract description 55
- 238000010606 normalization Methods 0.000 claims abstract description 24
- 238000005070 sampling Methods 0.000 claims description 65
- 230000006870 function Effects 0.000 claims description 22
- 238000004590 computer program Methods 0.000 claims description 12
- 230000004913 activation Effects 0.000 claims description 5
- 238000011176 pooling Methods 0.000 claims description 5
- 230000004807 localization Effects 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 15
- 230000008569 process Effects 0.000 description 15
- 238000004458 analytical method Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 230000006872 improvement Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000010224 classification analysis Methods 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000004148 unit process Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
- G06F3/015—Input arrangements based on nervous system activity detection, e.g. brain waves [EEG] detection, electromyograms [EMG] detection, electrodermal response detection
-
- 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
- A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
-
- 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
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The embodiment of the application provides an electroencephalogram signal band positioning method and device, and relates to the technical field of brain-computer interfaces, wherein the electroencephalogram signal band positioning method comprises the following steps: firstly, acquiring an electroencephalogram signal to be processed; normalization processing is carried out on the electroencephalogram signal to be processed to obtain a normalized signal; then processing the normalized signal through a pre-configured wave band positioning network to obtain output data; determining positioning data according to the output data and a preset confidence threshold; and finally, continuously processing the positioning information to remove noise in the positioning data to obtain an electroencephalogram signal band positioning result, so that the problem of band positioning of the electroencephalogram signal can be efficiently solved.
Description
Technical Field
The application relates to the field of brain-computer interfaces, in particular to a method and a device for positioning electroencephalogram signal wave bands.
Background
At present, electroencephalogram signal analysis is more and more widely applied to the field of brain-computer interfaces. Most of the existing electroencephalogram signal analysis still develops aiming at the electroencephalogram signal classification problem, namely, a fixed electroencephalogram signal is given, and the classification of the signal is judged by utilizing a classification method. For example, for the emotion analysis task, the signal is happy or sad; for the brain-computer interface task of moving the mechanical arm, whether the signal represents forward or backward needs to be given. The existing electroencephalogram signal classification method usually utilizes methods such as correlation, fractal dimension and the like, can only realize classification analysis on one section of signal, and cannot finish band positioning on the whole section of signal.
Disclosure of Invention
The embodiment of the application aims to provide an electroencephalogram signal band positioning method and device, and the problem of band positioning of electroencephalogram signals can be efficiently solved.
The first aspect of the embodiments of the present application provides a method for locating a band of an electroencephalogram signal, including: acquiring an electroencephalogram signal to be processed;
carrying out normalization processing on the electroencephalogram signal to be processed to obtain a normalized signal;
processing the normalized signal through a pre-configured wave band positioning network to obtain output data;
determining positioning data according to the output data and a preset confidence threshold;
and continuously processing the positioning information to remove noise in the positioning data and obtain an electroencephalogram signal wave band positioning result.
In the implementation process, firstly, electroencephalogram signals to be processed are obtained; normalization processing is carried out on the electroencephalogram signal to be processed to obtain a normalized signal; then processing the normalized signal through a pre-configured wave band positioning network to obtain output data; determining positioning data according to the output data and a preset confidence threshold; and finally, continuously processing the positioning information to remove noise in the positioning data to obtain an electroencephalogram signal band positioning result, and efficiently solving the problem of band positioning of the electroencephalogram signal.
Further, the determining the positioning data according to the output data and a preset confidence threshold includes:
acquiring a preset first confidence coefficient and a preset second confidence coefficient;
adjusting the confidence coefficient smaller than the confidence coefficient threshold value in the output data to the first confidence coefficient to obtain preliminary adjustment data; the output data comprises confidence corresponding to each sampling point;
and adjusting the confidence coefficient which is not less than the confidence coefficient threshold value in the preliminary adjustment data into the second confidence coefficient to obtain positioning data.
In the implementation process, threshold adjustment can be performed on output data through a preset first confidence coefficient and a preset second confidence coefficient, so that positioning data can be obtained.
Further, the continuously processing the positioning information to remove the noise in the positioning data and obtain the electroencephalogram signal band positioning result includes:
the first step is as follows: judging whether the confidence corresponding to one sampling point in the positioning data is the second confidence;
the second step is that: if the confidence coefficient is the second confidence coefficient, judging whether the confidence coefficient of other sampling points in the positioning data in a preset sampling range is the second confidence coefficient;
the third step: if the confidence coefficient does not exist, adjusting the confidence coefficient corresponding to the sampling point to be a first confidence coefficient;
the fourth step: traversing the confidence corresponding to each sampling point in the positioning data, and repeatedly executing the first step to the third step to finally obtain an electroencephalogram signal band positioning result.
In the implementation process, the positioning data is subjected to continuous processing, so that noise in the positioning data can be removed, and the improvement of the band positioning accuracy is facilitated.
Further, the band locating network comprises a CBL module, a CRP module, and a CRU module, wherein the CBL module comprises a first convolution sub-module, a normalization sub-module, and an activation function sub-module, the CRP module comprises the CBL module, a second convolution sub-module, and a pooling sub-module, and the CRU module comprises the CBL module, a third convolution sub-module, and an upsampling module.
Further, the band positioning network adopts a weighted cross entropy loss function during training, and the formula of the weighted cross entropy loss function is as follows:
WCE represents the weighted cross entropy loss function, n represents the number of sampling points corresponding to the output data, p is the confidence degree included by the output data, and betapRepresenting the confidence level in the output data as the ratio of the total number of the sampling points corresponding to the second confidence level to the total number of the sampling points corresponding to the first confidence level in the output data, wherein i is 1, 2, …, n, riAnd indicating the true value of the label corresponding to the ith sampling point.
Further, the normalizing the electroencephalogram signal to be processed to obtain a normalized signal includes:
calculating the maximum value of the amplitude of the electroencephalogram signal to be processed;
and dividing the real amplitude value of the electroencephalogram signal to be processed by the maximum amplitude value to obtain a normalized signal.
In the implementation process, the electroencephalogram signal to be processed is subjected to normalization processing, so that single-channel and one-dimensional data can be obtained, and the method is suitable for a pre-constructed band positioning network.
A second aspect of the embodiments of the present application provides an electroencephalogram signal band locating device, including:
the acquisition unit is used for acquiring an electroencephalogram signal to be processed;
the normalization unit is used for carrying out normalization processing on the electroencephalogram signal to be processed to obtain a normalized signal;
the model processing unit is used for processing the normalized signal through a pre-constructed wave band positioning network to obtain output data;
the determining unit is used for determining positioning data according to the output data and a preset confidence threshold;
and the continuity processing unit is used for carrying out continuity processing on the positioning information so as to remove noise in the positioning data and obtain an electroencephalogram signal band positioning result.
In the implementation process, the acquisition unit acquires an electroencephalogram signal to be processed; the normalization unit is used for performing normalization processing on the electroencephalogram signal to be processed to obtain a normalized signal; then the model processing unit processes the normalized signal through a pre-constructed band positioning network to obtain output data; the determining unit determines positioning data according to the output data and a preset confidence threshold; and finally, the continuity processing unit carries out continuity processing on the positioning information so as to remove noise in the positioning data, obtain an electroencephalogram signal band positioning result and effectively solve the problem of band positioning of the electroencephalogram signal.
Further, the determination unit includes:
the acquiring subunit is used for acquiring a preset first confidence coefficient and a preset second confidence coefficient;
the adjusting subunit is configured to adjust the confidence coefficient smaller than the confidence coefficient threshold in the output data to the first confidence coefficient, so as to obtain preliminary adjustment data; the output data comprises confidence corresponding to each signal sampling point; and adjusting the confidence coefficient which is not less than the confidence coefficient threshold value in the preliminary adjustment data into the second confidence coefficient to obtain positioning data.
In the implementation process, threshold adjustment can be performed on output data through a preset first confidence coefficient and a preset second confidence coefficient, so that positioning data can be obtained.
A third aspect of the embodiments of the present application provides an electronic device, including a memory and a processor, where the memory is used to store a computer program, and the processor runs the computer program to make the electronic device execute the electroencephalogram signal band locating method according to any one of the first aspect of the embodiments of the present application.
A fourth aspect of the present embodiment provides a computer-readable storage medium, which stores computer program instructions, where the computer program instructions, when read and executed by a processor, perform the electroencephalogram signal band localization method according to any one of the first aspect of the present embodiment.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments of the present application will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and that those skilled in the art can also obtain other related drawings based on the drawings without inventive efforts.
Fig. 1 is a schematic flowchart of a method for locating a band of an electroencephalogram signal according to an embodiment of the present application;
fig. 2 is a schematic flowchart of a method for locating a band of an electroencephalogram signal according to a second embodiment of the present application;
fig. 3 is a schematic structural diagram of an electroencephalogram signal band locating device provided in the third embodiment of the present application;
fig. 4 is a schematic structural diagram of an electroencephalogram signal band locating device provided in the fourth embodiment of the present application;
fig. 5 is a network diagram of a band locating network according to an embodiment of the present application;
fig. 6 is a schematic diagram of a positioning process of an electroencephalogram signal band according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of a band positioning network according to a second embodiment of the present application;
fig. 8 is a schematic structural diagram of a CBL module, a CRP module, and a CBL module according to the second embodiment of the present disclosure.
Detailed Description
The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
Example 1
Referring to fig. 1, fig. 1 is a schematic flow chart of a method for locating a band of an electroencephalogram signal according to an embodiment of the present application. The electroencephalogram signal wave band positioning method comprises the following steps:
s101, acquiring an electroencephalogram signal to be processed.
In the embodiment of the present application, the electroencephalogram signal to be processed may specifically be an original signal, and may also be an electroencephalogram signal that has undergone basic preprocessing such as filtering, power frequency interference removal, and the like, which is not limited in this embodiment of the present application.
S102, normalization processing is carried out on the electroencephalogram signal to be processed, and a normalized signal is obtained.
In the embodiment of the present application, in order to obtain data that can be suitable for a segmentation model (i.e., a preconfigured band positioning network), normalization processing needs to be performed on the data, and the data is integrally normalized within a certain fixed range, for example, between-1 and + 1. Because the signal amplitude is already processed to be close to 0 when baseline drift is removed, the amplitude range only needs to be limited, namely the maximum value of the amplitude data is solved, the maximum value of the absolute value is solved through the data in the data set, and finally the real value is divided by the maximum value to obtain the normalized data, so that the single-channel and one-dimensional data is obtained.
And S103, processing the normalized signal through a pre-constructed band positioning network to obtain output data.
In the embodiment of the present application, the preconfigured band locator network may specifically be a full convolution segmentation model, please refer to fig. 5, and fig. 5 is a network schematic diagram of a band locator network provided in the embodiment of the present application. As shown in fig. 5, the input terminal is a normalized signal, and the signal length of the input signal (i.e., the normalized signal) is variable, for example, the signal length of the normalized signal may be 1 × M, for example, when the oversampling frequency is 1000HZ and the sampling time duration is 60S, the signal length of the corresponding input signal is 1 × 60000, and this size is taken as an example in the following.
In the embodiment of the present application, the output size of the output data corresponds to the input, when the signal length of the input normalized signal is 1 × M, the size of the output data is also 1 × M, the size of the output data is the same as the input for the output without precision change, the signal length of the input signal is 1 × 60000 in fig. 5, the size of the output data is also 1 × 60000, and the output data includes the confidence corresponding to each sampling point.
And S104, determining positioning data according to the output data and a preset confidence threshold.
In the embodiment of the application, based on the output data, the electroencephalogram signal band can be located, and the specific process is as shown in fig. 6, because the output data includes the confidence corresponding to each sampling point, for a preset confidence threshold r (which may be set to 0.5), the confidence threshold r which is greater than or equal to the confidence threshold r is set to 1, and the confidence threshold r which is smaller than the confidence threshold r is set to 0; after the threshold processing is completed, the corresponding positioning data is obtained, and 5-10 and 16 parts shown in fig. 6 are the positioning data after the threshold processing.
And S105, continuously processing the positioning information to remove noise in the positioning data and obtain an electroencephalogram signal band positioning result.
In the embodiment of the present application, in an actual process, after the positioning data is obtained, prediction noise may occur in the positioning data, for example, 16 single points in fig. 6 obviously belong to the prediction noise, and therefore, a continuity process is required to remove noise in a prediction result (i.e., the positioning data).
In the embodiment of the present application, the specific method of the continuity processing is as follows: if the threshold processing result of the point i is 1, if the range from i-k to i + k has no result of setting 1, then the point i is set to 0, and meanwhile, the length of continuously setting 1 under the constraint of k range needs to be larger than a given length threshold L.
In the embodiment of the present application, the execution subject of the method may be a computing device such as a computer and a server, and is not limited in this embodiment.
In this embodiment, an execution subject of the method may also be an intelligent device such as a smart phone and a tablet computer, which is not limited in this embodiment.
Therefore, the electroencephalogram signal band positioning method described in the embodiment can efficiently solve the problem of band positioning of electroencephalogram signals.
Example 2
Please refer to fig. 2, fig. 2 is a schematic flow chart of a method for locating a band of an electroencephalogram signal according to an embodiment of the present application. As shown in fig. 2, the electroencephalogram signal band localization method includes:
s201, acquiring an electroencephalogram signal to be processed.
S202, calculating the maximum value of the amplitude of the electroencephalogram signal to be processed.
And S203, dividing the real amplitude value of the electroencephalogram signal to be processed by the maximum amplitude value to obtain a normalized signal.
In the embodiment of the present application, by implementing the steps S202 to S203, normalization processing can be performed on the electroencephalogram signal to be processed, so as to obtain a normalized signal.
In the embodiment of the application, in order to obtain data suitable for a segmentation model, normalization processing needs to be performed on the data, and the data is integrally normalized within a certain fixed range, such as a range from-1 to + 1. Because the signal amplitude is already processed to be close to 0 when baseline drift is removed, the amplitude range only needs to be limited, namely the maximum value of the amplitude data is solved, the maximum value of the absolute value is solved through the data in the data set, and finally the real value is divided by the maximum value to obtain the normalized data, so that the single-channel and one-dimensional data is obtained.
After step S203, the following steps are also included:
and S204, processing the normalized signal through a pre-constructed band positioning network to obtain output data.
Referring to fig. 7 and 8 together, fig. 7 is a schematic structural diagram of a band positioning network according to an embodiment of the present disclosure, and fig. 8 is a schematic structural diagram of a CBL module, a CRP module, and a CBL module according to an embodiment of the present disclosure. As shown in fig. 7 and 8, the band locator network includes a CBL module, a CRP module, and a CRU module, wherein the CBL module includes a first convolution sub-module, a normalization sub-module, and an activation function sub-module, the CRP module includes a CBL module, a second convolution sub-module, and a pooling sub-module, and the CRU module includes a CBL module, a third convolution sub-module, and an upsampling module.
In the above embodiment, the network core of the band positioning network includes two modules, namely, a CRP module and a CRU module, the CRP module is mainly used in the first half of the network, which includes convolution and other operations, and a pooling operation for down-sampling is included, so that the feature output dimension can be reduced; the latter half of the network is mainly based on a CRU module, and comprises an up-sampling module besides operations such as convolution and the like, wherein the up-sampling module can be realized by bilinear difference values, deconvolution and the like. Wherein the CRU module is capable of promoting the feature dimension, see fig. 5 for the reduction and promotion of the feature dimension.
In the above embodiments, the number of CRP modules and the number of CRU modules may be plural, and the specific number is not limited.
In the above embodiment, if the number of CRU modules is the same as the number of CRP modules, the output positioning accuracy is the same as the original data sampling resolution; if the number of CRU modules is less than that of CRP modules, the positioning precision with lower resolution can be realized; for example, a sampling rate of 1KHZ, if the number of CRU modules is the same as the number of CRP modules, the time resolution accuracy of 0.001 second is achieved; the temporal resolution will decrease by a factor of 2 for every few CRU modules.
In the embodiment of the present application, a specific structure of the preconfigured band locator network is shown in fig. 7, where the convolution represents a 1 × 3 one-dimensional convolution, BN represents BatchNorm, and the activation function may use leak Relu.
S205, acquiring a preset first confidence coefficient and a preset second confidence coefficient.
In this embodiment of the application, specifically, the preset first confidence may be 0, and the preset second confidence may be 1, which is not limited to this embodiment of the application.
As an alternative implementation, the band location network uses a weighted cross-entropy loss function during training, and the formula of the weighted cross-entropy loss function is:
WCE represents a weighted cross entropy loss function, n represents the number of sampling points corresponding to output data, p is confidence included in the output data, and betapThe ratio of the total number of the sampling points corresponding to the second confidence coefficient in the output data to the total number of the sampling points corresponding to the first confidence coefficient in the output data is represented, and i is 1, 2, …, n and riAnd indicating the true value of the label corresponding to the ith sampling point.
In the embodiment of the present application,
for example, when the first confidence is 0, the second confidence is 1,
in the embodiment of the application, for the ith sampling point, if the sampling point has a signal of a preset category, the tag true value corresponding to the sampling point is set as the second confidence, and if the sampling point does not have a signal of a preset category, the tag true value corresponding to the sampling point is set as the first confidence. Wherein, the preset category is a signal category and is preset. For example, when the first confidence is 0 and the second confidence is 1, for the ith sampling point, if the sampling point i has a signal in a preset category, the true label value corresponding to the sampling point i is set to 1, and if the sampling point i does not have a signal in a preset category, the true label value corresponding to the sampling point is set to 0.
In the embodiment of the present application, the signal category may be determined according to a mathematical relationship, a value characteristic, an energy power, a processing analysis, a time function characteristic, whether a value is a real number, or not.
After step S205, the following steps are also included:
s206, adjusting the confidence coefficient smaller than the confidence coefficient threshold value in the output data into a first confidence coefficient to obtain preliminary adjustment data; the output data includes the confidence level of each sample point.
And S207, adjusting the confidence coefficient which is not less than the confidence coefficient threshold value in the preliminary adjustment data into a second confidence coefficient to obtain the positioning data.
In the embodiment of the application, based on the output data, the electroencephalogram signal band can be located, and the specific process is as shown in fig. 6, because the output data includes the confidence corresponding to each sampling point, for a preset confidence threshold r (which may be set to 0.5), a preset second confidence is set for the confidence threshold r which is greater than or equal to the confidence threshold r, and a preset first confidence is set for the confidence threshold r which is smaller than the confidence threshold r; after the threshold processing is completed, the corresponding positioning data is obtained, and 5-10 and 16 parts shown in fig. 6 are the positioning data after the threshold processing.
In the embodiment of the present application, the positioning data can be determined according to the output data and the preset confidence threshold by implementing the steps S205 to S207.
After step S207, the following steps are also included:
and S208, continuously processing the positioning information to remove noise in the positioning data and obtain an electroencephalogram signal band positioning result.
As an optional implementation, the continuously processing the positioning information to remove noise in the positioning data to obtain a electroencephalogram signal band positioning result includes:
the first step is as follows: judging whether the confidence corresponding to one sampling point in the positioning data is a second confidence;
the second step is that: if the confidence coefficient is the second confidence coefficient, judging whether the confidence coefficient of other sampling points exists in the positioning data in the preset sampling range is the second confidence coefficient;
the third step: if the confidence coefficient does not exist, adjusting the confidence coefficient corresponding to the sampling point to be a first confidence coefficient;
the fourth step: traversing the confidence corresponding to each sampling point in the positioning data, and repeatedly executing the first step to the third step to finally obtain an electroencephalogram signal band positioning result.
In the above embodiment, the preset sampling range is preset, and may be specifically set to i-k to i + k, where i represents the sampling position of the currently processed sampling point.
In the above embodiment, when the first confidence is 0, the second confidence is the second confidence, and the preset sampling range is i-k to i + k, and the confidence corresponding to the i point of one of the sampling points i in the positioning data is the second confidence, if there is no result of setting 1 in the range from i-k to i + k, then setting the i point to 0, and meanwhile, the length of continuously setting 1 under the constraint of the k range needs to be greater than the given length threshold L.
Therefore, the electroencephalogram signal band positioning method described in the embodiment can efficiently solve the problem of band positioning of electroencephalogram signals.
Example 3
Please refer to fig. 3, fig. 3 is a schematic structural diagram of an electroencephalogram signal band locating device according to an embodiment of the present application. As shown in fig. 3, the electroencephalogram signal band locating device includes:
an obtaining unit 310, configured to obtain an electroencephalogram signal to be processed;
the normalization unit 320 is configured to perform normalization processing on the electroencephalogram signal to be processed to obtain a normalized signal;
the model processing unit 330 is configured to process the normalized signal through a preconfigured band positioning network to obtain output data;
a determining unit 340, configured to determine positioning data according to the output data and a preset confidence threshold;
and the continuity processing unit 350 is configured to perform continuity processing on the positioning information to remove noise in the positioning data, so as to obtain a positioning result of the electroencephalogram signal band.
In the embodiment of the present application, for the explanation of the electroencephalogram signal band locating apparatus, reference may be made to the description in embodiment 1 or embodiment 2, and details are not repeated in this embodiment.
Therefore, the electroencephalogram signal band positioning device described in the embodiment can efficiently solve the problem of band positioning of electroencephalogram signals.
Example 4
Referring to fig. 4, fig. 4 is a schematic structural diagram of an electroencephalogram signal band locating device according to an embodiment of the present application. The electroencephalogram signal band locating device shown in fig. 4 is obtained by optimizing the electroencephalogram signal band locating device shown in fig. 3. As shown in fig. 4, the determination unit 340 includes:
an obtaining subunit 341, configured to obtain a preset first confidence level and a preset second confidence level;
an adjusting subunit 342, configured to adjust a confidence level that is smaller than the confidence level threshold in the output data to a first confidence level, so as to obtain preliminary adjustment data; the output data comprises confidence corresponding to each signal sampling point; and adjusting the confidence coefficient which is not less than the confidence coefficient threshold value in the preliminary adjustment data into a second confidence coefficient to obtain the positioning data.
As an alternative embodiment, the continuity processing unit 350 includes:
the first judging subunit 351 is configured to judge whether a confidence corresponding to one of the sampling points in the positioning data is a second confidence;
the second determining subunit 352 is configured to determine, when the first determining subunit 351 determines that the confidence is the second confidence, whether the confidence of other sampling points in the positioning data in the preset sampling range is the second confidence;
an adjusting subunit 353, configured to adjust the confidence level corresponding to the sampling point to a first confidence level when the second determining subunit 352 determines that the sampling point does not exist;
the circulation subunit 354 is configured to traverse the confidence corresponding to each sampling point in the positioning data, and repeatedly trigger the first determining subunit 351, the second determining subunit 352, and the adjusting subunit 353 to execute corresponding operations, so as to finally obtain a electroencephalogram signal band positioning result.
As an optional implementation manner, the band locator network includes a CBL module, a CRP module, and a CRU module, where the CBL module includes a first convolution sub-module, a normalization sub-module, and an activation function sub-module, the CRP module includes a CBL module, a second convolution sub-module, and a pooling sub-module, and the CRU module includes a CBL module, a third convolution sub-module, and an upsampling module.
As an optional implementation, the band location network uses a weighted cross-entropy loss function during training, and the formula of the weighted cross-entropy loss function is:
WCE represents the weighted cross entropy loss function, n represents the number of sampling points corresponding to the output data, p is the confidence degree included by the output data, and betapRepresenting the ratio of the total number of the sampling points corresponding to the second confidence coefficient in the output data to the total number of the sampling points corresponding to the first confidence coefficient in the output data, wherein i is 1, 2, …, n, riAnd indicating the true value of the label corresponding to the ith sampling point.
As an optional implementation, the normalization unit 320 includes:
the first calculating subunit 321 is configured to calculate a maximum amplitude value of the electroencephalogram signal to be processed;
and the second calculating subunit 322 is configured to divide the actual amplitude value of the electroencephalogram signal to be processed by the maximum amplitude value to obtain a normalized signal.
In the embodiment of the present application, for the explanation of the electroencephalogram signal band locating device, reference may be made to the description in embodiment 1 or embodiment 2, and details are not repeated in this embodiment.
Therefore, the electroencephalogram signal band positioning device described in the embodiment can efficiently solve the problem of band positioning of electroencephalogram signals.
The embodiment of the application provides an electronic device, which comprises a memory and a processor, wherein the memory is used for storing a computer program, and the processor runs the computer program to enable the electronic device to execute the electroencephalogram signal band locating method in any one of embodiment 1 and embodiment 2 of the application.
The embodiment of the present application provides a computer-readable storage medium, which stores computer program instructions, and when the computer program instructions are read and executed by a processor, the method for locating a band of an electroencephalogram signal according to any one of embodiments 1 and 2 of the present application is executed.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above description is only an example of the present application and is not intended to limit the scope of the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Claims (10)
1. An electroencephalogram signal band positioning method is characterized by comprising the following steps:
acquiring an electroencephalogram signal to be processed;
carrying out normalization processing on the electroencephalogram signal to be processed to obtain a normalized signal;
processing the normalized signal through a pre-configured wave band positioning network to obtain output data;
determining positioning data according to the output data and a preset confidence threshold;
and continuously processing the positioning information to remove noise in the positioning data and obtain an electroencephalogram signal wave band positioning result.
2. The method for locating the bands of electroencephalogram signals according to claim 1, wherein the determining the location data according to the output data and a preset confidence threshold comprises:
acquiring a preset first confidence coefficient and a preset second confidence coefficient;
adjusting the confidence coefficient smaller than the confidence coefficient threshold value in the output data to the first confidence coefficient to obtain preliminary adjustment data; the output data comprises confidence corresponding to each sampling point;
and adjusting the confidence coefficient which is not less than the confidence coefficient threshold value in the preliminary adjustment data into the second confidence coefficient to obtain positioning data.
3. The method for locating the band of electroencephalogram signal according to claim 2, wherein the continuously processing the locating information to remove the noise in the locating data and obtain the result of locating the band of electroencephalogram signal comprises:
the first step is as follows: judging whether the confidence corresponding to one sampling point in the positioning data is the second confidence;
the second step: if the confidence coefficient is the second confidence coefficient, judging whether the confidence coefficient of other sampling points in the positioning data in a preset sampling range is the second confidence coefficient;
the third step: if the confidence coefficient does not exist, adjusting the confidence coefficient corresponding to the sampling point to be a first confidence coefficient;
the fourth step: traversing the confidence corresponding to each sampling point in the positioning data, and repeatedly executing the first step to the third step to finally obtain an electroencephalogram signal band positioning result.
4. The method of claim 1, wherein the band locator network comprises a CBL module, a CRP module, and a CRU module, wherein the CBL module comprises a first convolution sub-module, a normalization sub-module, and an activation function sub-module, the CRP module comprises the CBL module, a second convolution sub-module, and a pooling sub-module, and the CRU module comprises the CBL module, a third convolution sub-module, and an upsampling module.
5. The electroencephalogram signal band location method of claim 2, wherein the band location network adopts a weighted cross-entropy loss function during training, and the formula of the weighted cross-entropy loss function is as follows:
wherein WCE represents the weighted cross entropy loss function, n represents the number of sampling points corresponding to the output data, p is the confidence included in the output data, and βpRepresenting that the confidence level in the output data is the total number of the sampling points corresponding to the second confidence level and the confidence level in the output data is the first confidence levelConfidence corresponds to the ratio between the total number of sample points, i ═ 1, 2, …, n, riAnd indicating the true value of the label corresponding to the ith sampling point.
6. The electroencephalogram signal band positioning method according to claim 1, wherein the normalization processing is performed on the electroencephalogram signal to be processed to obtain a normalized signal, and the normalization processing comprises:
calculating the maximum value of the amplitude of the electroencephalogram signal to be processed;
and dividing the real amplitude value of the electroencephalogram signal to be processed by the maximum amplitude value to obtain a normalized signal.
7. An electroencephalogram signal band locating device, characterized in that the electroencephalogram signal band locating device comprises:
the acquisition unit is used for acquiring an electroencephalogram signal to be processed;
the normalization unit is used for performing normalization processing on the electroencephalogram signal to be processed to obtain a normalized signal;
the model processing unit is used for processing the normalized signal through a pre-constructed wave band positioning network to obtain output data;
the determining unit is used for determining positioning data according to the output data and a preset confidence threshold;
and the continuity processing unit is used for carrying out continuity processing on the positioning information so as to remove noise in the positioning data and obtain an electroencephalogram signal band positioning result.
8. The electroencephalogram signal band locating apparatus according to claim 7, wherein the determining unit includes:
the acquiring subunit is used for acquiring a preset first confidence coefficient and a preset second confidence coefficient;
the adjusting subunit is configured to adjust the confidence coefficient smaller than the confidence coefficient threshold in the output data to the first confidence coefficient, so as to obtain preliminary adjustment data; the output data comprises confidence corresponding to each signal sampling point; and adjusting the confidence coefficient which is not less than the confidence coefficient threshold value in the preliminary adjustment data into the second confidence coefficient to obtain positioning data.
9. An electronic device, characterized in that the electronic device comprises a memory for storing a computer program and a processor for executing the computer program to cause the electronic device to perform the electroencephalogram signal band localization method of any one of claims 1 to 6.
10. A readable storage medium having stored thereon computer program instructions which, when read and executed by a processor, perform the method of electroencephalogram signal band localization according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111111846.9A CN113791691B (en) | 2021-09-18 | 2021-09-18 | Electroencephalogram signal band positioning method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111111846.9A CN113791691B (en) | 2021-09-18 | 2021-09-18 | Electroencephalogram signal band positioning method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113791691A CN113791691A (en) | 2021-12-14 |
| CN113791691B true CN113791691B (en) | 2022-05-20 |
Family
ID=79184123
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111111846.9A Active CN113791691B (en) | 2021-09-18 | 2021-09-18 | Electroencephalogram signal band positioning method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113791691B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4083365A (en) * | 1976-06-10 | 1978-04-11 | Don Robert Yancey | Dual integrator EEG analyzer |
| US5715452A (en) * | 1993-12-27 | 1998-02-03 | Hitachi, Ltd. | Process of transferring file, process of gaining access to data and process of writing data |
| CN109247917A (en) * | 2018-11-21 | 2019-01-22 | 广州大学 | A kind of spatial hearing induces P300 EEG signal identification method and device |
| CN110069958A (en) * | 2018-01-22 | 2019-07-30 | 北京航空航天大学 | A kind of EEG signals method for quickly identifying of dense depth convolutional neural networks |
| CN110353702A (en) * | 2019-07-02 | 2019-10-22 | 华南理工大学 | A kind of emotion identification method and system based on shallow-layer convolutional neural networks |
| CN112070111A (en) * | 2020-07-28 | 2020-12-11 | 浙江大学 | Multi-target detection method and system adaptive to multiband images |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140020089A1 (en) * | 2012-07-13 | 2014-01-16 | II Remo Peter Perini | Access Control System using Stimulus Evoked Cognitive Response |
| US11717686B2 (en) * | 2017-12-04 | 2023-08-08 | Neuroenhancement Lab, LLC | Method and apparatus for neuroenhancement to facilitate learning and performance |
-
2021
- 2021-09-18 CN CN202111111846.9A patent/CN113791691B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4083365A (en) * | 1976-06-10 | 1978-04-11 | Don Robert Yancey | Dual integrator EEG analyzer |
| US5715452A (en) * | 1993-12-27 | 1998-02-03 | Hitachi, Ltd. | Process of transferring file, process of gaining access to data and process of writing data |
| CN110069958A (en) * | 2018-01-22 | 2019-07-30 | 北京航空航天大学 | A kind of EEG signals method for quickly identifying of dense depth convolutional neural networks |
| CN109247917A (en) * | 2018-11-21 | 2019-01-22 | 广州大学 | A kind of spatial hearing induces P300 EEG signal identification method and device |
| CN110353702A (en) * | 2019-07-02 | 2019-10-22 | 华南理工大学 | A kind of emotion identification method and system based on shallow-layer convolutional neural networks |
| CN112070111A (en) * | 2020-07-28 | 2020-12-11 | 浙江大学 | Multi-target detection method and system adaptive to multiband images |
Non-Patent Citations (2)
| Title |
|---|
| 基于FastICA和卷积神经网络的脑电信号分类算法;陈宇等;《黑龙江大学自然科学学报》;20180625(第03期);全文 * |
| 基于时频域分析的运动想象脑电信号分类;穆振东等;《中国组织工程研究与临床康复》;20090625(第26期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113791691A (en) | 2021-12-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6765445B2 (en) | Frequency-based audio analysis using neural networks | |
| CN110246490B (en) | Voice keyword detection method and related device | |
| CN109344789B (en) | Face tracking method and device | |
| CN113870334B (en) | Depth detection method, device, equipment and storage medium | |
| CN106294331A (en) | Audio information retrieval method and device | |
| CN110718228A (en) | Voice separation method and device, electronic equipment and computer readable storage medium | |
| CN111429932A (en) | Voice noise reduction method, device, equipment and medium | |
| CN107680584B (en) | Method and device for segmenting audio | |
| CN113284001A (en) | Power consumption prediction method and device, computer equipment and storage medium | |
| Wang et al. | High quality impulse noise removal via non‐uniform sampling and autoregressive modelling based super‐resolution | |
| CN114820351A (en) | Hyperspectral noise filtering method, device, equipment and storage medium | |
| CN113791691B (en) | Electroencephalogram signal band positioning method and device | |
| CN107894837A (en) | Dynamic sentiment analysis model sample processing method and processing device | |
| CN111414993A (en) | Cutting and convolution calculating method and device of convolution neural network | |
| CN113688655B (en) | Method, device, computer equipment and storage medium for identifying interference signals | |
| CN113139561B (en) | Garbage classification method, garbage classification device, terminal equipment and storage medium | |
| CN113392902A (en) | Data set processing method and device, storage medium and electronic equipment | |
| CN114882334A (en) | Method for generating pre-training model, model training method and device | |
| CN105513587B (en) | MFCC extraction method and device | |
| CN113724720A (en) | Non-human voice filtering method in noisy environment based on neural network and MFCC | |
| CN111090723A (en) | Power grid safety production content recommendation method based on knowledge graph | |
| Goh et al. | Robust speech recognition system using multi-parameter bidirectional Kalman filter | |
| CN110888986B (en) | Information pushing method, device, electronic equipment and computer readable storage medium | |
| CN115035911B (en) | Noise generation model training method, device, equipment and medium | |
| CN115511015B (en) | Sample screening method, device, equipment and computer readable storage medium |
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 |