CN114403899A - Depression detection method combining cerebral neuron spike potential and local field potential - Google Patents
Depression detection method combining cerebral neuron spike potential and local field potential Download PDFInfo
- Publication number
- CN114403899A CN114403899A CN202210117132.7A CN202210117132A CN114403899A CN 114403899 A CN114403899 A CN 114403899A CN 202210117132 A CN202210117132 A CN 202210117132A CN 114403899 A CN114403899 A CN 114403899A
- Authority
- CN
- China
- Prior art keywords
- data
- learning network
- ensemble learning
- target object
- electroencephalogram
- 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
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/369—Electroencephalography [EEG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/165—Evaluating the state of mind, e.g. depression, anxiety
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Abstract
The invention discloses a depression detection method combining cerebral neuron spike potential and local field potential, which comprises the steps of respectively obtaining electroencephalogram data of a target object in a depressed state and a non-depressed state; performing feature fusion on brain neuron spike potential data and local field potential data corresponding to the same electroencephalogram data to obtain fused electroencephalogram data; training an ensemble learning network based on training data, determining optimal network parameters, and optimizing the ensemble learning network based on the optimal network parameters; and performing classification prediction on the current electroencephalogram data of the target object based on an ensemble learning network, performing voting selection on each output classification prediction result, and determining the depression state of the target object based on the final classification result. The invention realizes the purpose of comprehensively obtaining the final prediction result by combining the advantages of the integrated learning and different classifiers and voting and selecting the results of the three classifiers. The characteristics in the signals can be effectively extracted, and the depression state can be accurately judged.
Description
Technical Field
The application relates to the technical field of electroencephalogram data analysis, in particular to a depression detection method combining cerebral neuron spike potential and local field potential.
Background
The electroencephalogram data signal is an information carrier carrying brain states and is an important tool for judging the brain states. Because electroencephalogram data is usually a complex signal at a high latitude, effective information in the electroencephalogram signal is usually required to be acquired through an electroencephalogram signal feature extraction technology. Based on the electroencephalogram signal extraction technology, the emotional state of the brain can be analyzed, and the method has important research and clinical application significance for diagnosis and treatment of cranial nerve system diseases such as depression and epilepsy.
The invention with the application number of 202111049290.5 discloses an electroencephalogram signal analysis method, which adopts a multiband time-space convolution network to analyze extracted features, effectively extracts features related to depression from the characteristics of the electroencephalogram signal through the time-space convolution network, and realizes classification of the depression electroencephalogram.
The invention with the application number of 200910196746.3 discloses an electroencephalogram analysis method, which extracts electroencephalogram signal characteristics by applying time domain and frequency domain analysis and principal component analysis methods, extracts time-frequency domain parameters related to states of human bodies such as tension, fatigue and relaxation, maps the time-frequency domain parameters into a principal component space, further applies a support vector machine to analyze nonlinear relations in the principal component space, and improves the accuracy and the effectiveness of interpretation.
Although the method can extract the electroencephalogram characteristics to a certain extent, for depression with complex electroencephalogram signals, the method cannot well utilize information in the electroencephalogram signals, so that the problems that part of key information in the electroencephalogram signals is lost or the extracted characteristics are invalid in state judgment of the depression exist.
Disclosure of Invention
In order to solve the above problems, the embodiments of the present application provide a method for detecting depression by combining cerebral neuron spikes with local field potentials.
In a first aspect, the present embodiments provide a method for detecting depression by combining cerebral neuron spikes with local field potentials, the method comprising:
acquiring electroencephalogram data of a target object in a depressed state and a non-depressed state respectively, wherein the electroencephalogram data comprise cerebral neuron spike potential data and local field potential data;
performing feature fusion on the brain neuron spike potential data and the local field potential data corresponding to the same electroencephalogram data to obtain fused electroencephalogram data, and taking the fused electroencephalogram data as training data;
establishing an ensemble learning network, training the ensemble learning network based on the training data, determining optimal network parameters, and optimizing the ensemble learning network based on the optimal network parameters;
classifying and predicting the current electroencephalogram data of the target object based on the ensemble learning network to obtain at least one output classification prediction result, voting and selecting each output classification prediction result to obtain a final classification result, and determining the depression state of the target object based on the final classification result.
Preferably, the performing feature fusion on the brain neuron spike potential data and the local field potential data corresponding to the same electroencephalogram data to obtain fused electroencephalogram data includes:
extracting the information of the number of issues of the cerebral neuron spike potential data, extracting the power spectrum information of the local field potential data, and then respectively normalizing the information of the number of issues and the power spectrum information;
and performing characteristic fusion on the release number information and the power spectrum information corresponding to the same electroencephalogram data through data splicing to obtain fused electroencephalogram data.
Preferably, the extracting the information on the number of issues of the cerebral neuron spike potential data and the extracting the power spectrum information of the local field potential data includes:
acquiring the total issuing number of neurons in preset time duration from the brain neuron spike potential data through a sliding window, wherein the total issuing number is issuing number information;
and acquiring the frequency spectrum density in a preset time length in the local field potential data through the sliding window, wherein the frequency spectrum density is power spectrum information.
Preferably, the using the fused electroencephalogram data as training data includes:
randomly dividing the fusion electroencephalogram data into training data and testing data;
after optimizing the ensemble learning network based on the optimal network parameters, the method further includes:
validating the ensemble learning network based on the test data.
Preferably, the establishing an ensemble learning network, training the ensemble learning network based on the training data, determining an optimal network parameter, and optimizing the ensemble learning network based on the optimal network parameter includes:
establishing an ensemble learning network, wherein a classifier in the ensemble learning network comprises a support vector machine, a k-nearest neighbor clustering and a width learning neural network;
training the ensemble learning network based on the training data, respectively determining optimal network parameters corresponding to the support vector machine, the k-nearest neighbor clustering and the width learning neural network, and optimizing the ensemble learning network based on each optimal network parameter.
Preferably, the classifying and predicting the current electroencephalogram data of the target object based on the ensemble learning network to obtain at least one output classification prediction result, voting and selecting each output classification prediction result to obtain a final classification result, and determining the depression state of the target object based on the final classification result includes:
based on each classifier of the ensemble learning network, performing classification prediction calculation on the current electroencephalogram data of the target object respectively to obtain three output classification prediction results;
voting and selecting each output classification prediction result to obtain a final classification result;
determining a depression status of the target subject based on the final classification result.
Preferably, the method further comprises:
constructing a mapping relation between a target object and the integrated learning network;
and when the new target object is determined, searching the ensemble learning network corresponding to the target object based on the mapping relation.
In a second aspect, embodiments of the present application provide a depression detection apparatus in which cerebral neuron spikes are combined with local field potentials, the apparatus comprising:
the acquisition module is used for respectively acquiring electroencephalogram data of a target object in a depression state and a non-depression state, wherein the electroencephalogram data comprise cerebral neuron spike potential data and local field potential data;
the fusion module is used for performing feature fusion on the brain neuron spike potential data and the local field potential data corresponding to the same electroencephalogram data to obtain fused electroencephalogram data, and taking the fused electroencephalogram data as training data;
the building module is used for building an ensemble learning network, training the ensemble learning network based on the training data, determining an optimal network parameter, and optimizing the ensemble learning network based on the optimal network parameter;
the prediction module is used for carrying out classification prediction on the current electroencephalogram data of the target object based on the ensemble learning network to obtain at least one output classification prediction result, voting and selecting each output classification prediction result to obtain a final classification result, and determining the depression state of the target object based on the final classification result.
In a third aspect, an embodiment of the present application provides an electronic device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the computer program to implement the steps of the method as provided in the first aspect or any one of the possible implementation manners of the first aspect.
In a fourth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the method as provided in the first aspect or any one of the possible implementations of the first aspect.
The invention has the beneficial effects that: the comprehensive application of the brain nerve signal peak potential signal and the local field potential is considered, the two signals are fused, the signal characteristics more effective than the brain nerve signals such as cortical nerve signals are obtained, and three classifiers are designed by utilizing the advantages of integrated learning and combining different classifiers: the method comprises the steps of supporting a vector machine, k-nearest neighbor and a width learning network, combining the advantages of a linear classifier, a clustering algorithm and the learning network, and then adopting a minority majority-obeying voting principle to the results of the linear classifier, the clustering algorithm and the learning network to comprehensively obtain a final prediction result. The characteristics in the signals can be effectively extracted, the current state of the depression patient can be accurately judged, and then the treatment of the patient by the auxiliary treatment machine can be assisted.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for detecting depression by combining cerebral neuron spikes with local field potentials according to an embodiment of the present disclosure;
FIG. 2 is a schematic diagram of a method for detecting depression by combining brain neuron spikes with local field potentials according to an embodiment of the present disclosure;
FIG. 3 is a schematic structural diagram of a device for detecting depression by combining cerebral neuron spikes with local field potentials according to an embodiment of the present disclosure;
fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.
In the following description, the terms "first" and "second" are used for descriptive purposes only and are not intended to indicate or imply relative importance. The following description provides embodiments of the present application, where different embodiments may be substituted or combined, and thus the present application is intended to include all possible combinations of the same and/or different embodiments described. Thus, if one embodiment includes feature A, B, C and another embodiment includes feature B, D, then this application should also be considered to include an embodiment that includes one or more of all other possible combinations of A, B, C, D, even though this embodiment may not be explicitly recited in text below.
The following description provides examples, and does not limit the scope, applicability, or examples set forth in the claims. Changes may be made in the function and arrangement of elements described without departing from the scope of the disclosure. Various examples may omit, substitute, or add various procedures or components as appropriate. For example, the described methods may be performed in an order different than the order described, and various steps may be added, omitted, or combined. Furthermore, features described with respect to some examples may be combined into other examples.
Referring to fig. 1, fig. 1 is a schematic flow chart of a depression detection method for combining cerebral neuron spikes with local field potentials according to an embodiment of the present application. In an embodiment of the present application, the method includes:
s101, acquiring electroencephalogram data of a target object in a depression state and a non-depression state respectively, wherein the electroencephalogram data comprise cerebral neuron spike potential data and local field potential data.
The execution main body of the application can be a server of a computer terminal.
In the embodiment of the application, the electroencephalogram signals can change along with the change of the self state of an individual and the surrounding environment, and the electroencephalogram signals of a depressive individual and a normal individual have a certain difference, so that the electroencephalogram signals are subjected to feature extraction from the spike potential and the local field potential of cerebral neurons and the depressive state is predicted. Illustratively, the data employed herein may be invasive cranial nerve signals from mice labeled for a depressed state with a recording duration of 600s for the depressed state and 3600s for the non-depressed state. The spike potential electroencephalogram signals and the local field potential electroencephalogram signals of the cerebral neurons are acquired simultaneously in the acquisition process.
S102, performing feature fusion on the brain neuron spike potential data and the local field potential data corresponding to the same electroencephalogram data to obtain fused electroencephalogram data, and taking the fused electroencephalogram data as training data.
In the embodiment of the present application, as shown in fig. 2, for the electroencephalogram data acquired at the same time, feature fusion processing is performed on the corresponding cerebral neuron spike data and the local field potential data, so as to obtain fused electroencephalogram data, and the fused electroencephalogram data can be used as subsequent training data.
In one possible embodiment, step S102 includes:
extracting the information of the number of issues of the cerebral neuron spike potential data, extracting the power spectrum information of the local field potential data, and then respectively normalizing the information of the number of issues and the power spectrum information;
and performing characteristic fusion on the release number information and the power spectrum information corresponding to the same electroencephalogram data through data splicing to obtain fused electroencephalogram data.
In the embodiment of the application, the information of the number of issues is extracted from the brain neuron spike data, the power spectrum information is extracted from the local field potential data, then the server performs normalization processing based on the maximum and minimum values on the extracted characteristic information, and performs characteristic fusion on the two normalized information data, so as to obtain the fused electroencephalogram data.
Illustratively, for the peak potential data S, the dimension is t ×
Where t is the total number of time steps,
the number of neurons collected for peak potentials. For the local field potential data L, the dimension is t c m, where t is the total number of time steps, which is the same as t in the peak potential; c is the number of channels of the local field potential, and m is the number of frequency bands for extracting the frequency spectrum density division. Converting the dimensionality of local field potential data into t
Wherein
. Connecting the merged peak potential data S and the local field potential data L to obtain fused electroencephalogram data F with the dimensionality t
Wherein
。
Specifically, the dimension of the spike data for the depressed state is 239 × 17, and the dimension of the spike data for the non-depressed state is 1439 × 17; the dimension of the local field potential data for the depressed state was 239 × 16 × 7, and after the dimension conversion was 239 × 112, and the dimension of the local field potential data for the non-depressed state was 1439 × 16 × 7, and after the dimension conversion was 1439 × 112. After signal fusion, the dimension of the fused signal data in the depressed state is 239 × 129, and the dimension of the fused signal data in the non-depressed state is 1439 × 129.
In one embodiment, the extracting the information on the number of issues of the cerebral neuron spike data and the extracting the information on the power spectrum of the local field potential data includes:
acquiring the total issuing number of neurons in preset time duration from the brain neuron spike potential data through a sliding window, wherein the total issuing number is issuing number information;
and acquiring the frequency spectrum density in a preset time length in the local field potential data through the sliding window, wherein the frequency spectrum density is power spectrum information.
In the embodiment of the application, the server obtains the total number of the neurons in 5s by using the brain neuron spike data through a sliding window method, wherein the overlapping percentage of the sliding window is 50%, namely the step length is 2.5 s; extracting the frequency spectrum density of the local field potential data according to a time window of 5s, wherein the coincidence percentage of the sliding window is 50%, namely the step length is 2.5s, and the frequency spectrum comprises a delta wave band of 0.5-4Hz, a theta wave band of 4-8Hz, an alpha wave band of 8-13Hz, a beta wave band of 13-30Hz, a gamma wave band of 30-70Hz, a high gamma wave band of 70-150Hz and an extremely high gamma wave band of 150-250 Hz.
In one embodiment, the using the fused electroencephalogram data as training data includes:
dividing the fused electroencephalogram data into training data and testing data;
after optimizing the ensemble learning network based on the optimal network parameters, the method further includes:
validating the ensemble learning network based on the test data.
In the embodiment of the application, the fused electroencephalogram data can be divided into training data and testing data, and after the network is trained through a training model, in order to ensure the accuracy of the network in actual use, the integrated learning network is verified through the testing data. Specifically, when a training set and a test set are divided, layered sampling is performed according to a depressive state and a non-depressive state, and the proportion of the training set to the test set is 7: 3, the training set fusion signal data has a dimension of 1174 x 129, and the test set fusion signal data has a dimension of 504 x 129.
S103, establishing an ensemble learning network, training the ensemble learning network based on the training data, determining an optimal network parameter, and optimizing the ensemble learning network based on the optimal network parameter.
In the embodiment of the application, the server establishes the ensemble learning network, so that the ensemble learning network is trained and learned through training data, the optimal network parameters of all parts of the ensemble learning network are determined in the learning process, and finally the ensemble learning network is reversely optimized according to the optimal network parameters determined by training.
In one possible embodiment, step S103 includes:
establishing an ensemble learning network, wherein a classifier in the ensemble learning network comprises a support vector machine, a k-nearest neighbor clustering and a width learning neural network;
training the ensemble learning network based on the training data, respectively determining optimal network parameters corresponding to the support vector machine, the k-nearest neighbor clustering and the width learning neural network, and optimizing the ensemble learning network based on each optimal network parameter.
In the embodiment of the application, after the ensemble learning network is established, the training data is utilized to train the ensemble learning network to obtain the optimal parameters of each part of the network. The integrated network comprises a support vector machine, k-nearest neighbor clustering and a width learning neural network. The network principles and parameters of the various parts are as follows:
and a support vector machine. The method comprises the steps of searching performance optimal parameters in a parameter grid searching mode, wherein parameters adjusted by a support vector machine through training are { C, gamma and kernel }, wherein C is the punishment degree of classification errors, gamma is the coefficient of a kernel function, and kernel is the type of the kernel function. The optimal parameters on the spike potential are { C: 1000, gamma: 0.1, kernel: 'rbf', the optimal parameters at the local field potential are { C: 1000, gamma: 0.1, kernel: 'rbf', the optimal parameters on the fusion signal are { C: 100, gamma: 0.1, kernel: 'rbf'.
And k-nearest neighbor clustering. And performing performance optimal parameter search in a parameter grid search mode, wherein parameters of k-neighbor clustering adjusted through training have { n _ neighbors }, wherein the n _ neighbors are clustering numbers of a k-neighbor clustering algorithm. The optimal parameters on the spike potential are { n _ neighbors: 5, the optimal parameters at the local field potential are { n _ neighbors: 5, the optimal parameters on the fusion signal are { n _ neighbors: 3}.
A width learning network. For the fusion electroencephalogram data F, the corresponding depression state is Y, and a weight matrix is randomly generated
And offset
Performing feature mapping on the electroencephalogram data to obtain
The following relationship is satisfied:
where Relu is a linear rectification function. Repetition of
Then obtain
Group mapping of characteristic data, applying this
Data of group mapping characteristics is composed of time-aligned new data
Randomly generating a weight matrix
And offset
Orthogonalizing the weight matrix to obtain a new weight matrix
Will countAccording to
Mapping to obtain new data
The following relationship is satisfied:
repetition of
Then, obtain
Group of
Mapped data, which is then processed
Data of group mapping characteristics is composed of time-aligned new data
Will be
And
obtaining new EEG mapped data along time
Initializing a weight
With predicted results
Satisfy the following relationships：
To obtain weights at optimal performance
Then, the following is obtained:
wherein
Is a lagrange multiplier.
Then, the least squares method is used to solve the optimum
：
Optimal network parameters using training set
In combination with manually specified hyper-parameters
，
，
And finding the optimal hyper-parameter set on the training set through parameter search to finish the training of the network.
S104, performing classification prediction on the current electroencephalogram data of the target object based on the ensemble learning network to obtain at least one output classification prediction result, performing voting selection on each output classification prediction result to obtain a final classification result, and determining the depression state of the target object based on the final classification result.
In the embodiment of the application, after the ensemble learning network is trained to predict the target object, the actual current electroencephalogram data of the target object is classified and predicted based on the ensemble learning network, and at least one output classification prediction result is obtained through respective calculation of each part in the ensemble learning network. And finally, voting and selecting each output classification prediction result to determine a final classification result, and further determining the current depression state of the target object.
It is emphasized that the method of the present application is not intended for the diagnosis or treatment of depression, but rather for the detection and determination of the current state of a patient who has been diagnosed with depression, whereby the auxiliary treatment machine is able to treat the patient at the right time.
In one possible embodiment, step S104 includes:
based on each classifier of the ensemble learning network, performing classification prediction calculation on the current electroencephalogram data of the target object respectively to obtain three output classification prediction results;
voting and selecting each output classification prediction result to obtain a final classification result;
determining a depression status of the target subject based on the final classification result.
In the embodiment of the application, after each part of the integrated network is trained by using training data, the prediction results of the three parts of the integrated network are voted. Specifically, the voting choice is calculated as follows:
wherein the content of the first and second substances,
for integrating the netThe prediction results of the support vector machines in the network,
to integrate the prediction results of K-neighbors in the network,
for prediction results of a breadth-learning network in an integrated network "
"means the whole division by the number of times,
indicating the final prediction result, with a prediction result of 1 indicating that the individual is in a depressed state and a prediction result of 0 indicating that the individual is in a non-depressed state for each time period.
In one embodiment, the method further comprises:
constructing a mapping relation between a target object and the integrated learning network;
and when the new target object is determined, searching the ensemble learning network corresponding to the target object based on the mapping relation.
In the embodiment of the application, after the ensemble learning network is trained, the mapping relationship between the target object and the ensemble learning network is constructed, so that when a new target object needing depression prediction appears, the corresponding ensemble learning network is directly found according to the mapping relationship, and the data processing efficiency is improved.
The depression detecting device for binding brain neuron spikes with local field potentials provided by the embodiments of the present application will be described in detail with reference to fig. 3. It should be noted that the depression detecting apparatus shown in fig. 3 is used for performing the method of the embodiment shown in fig. 1 of the present application, and for convenience of illustration, only the relevant portions of the embodiment of the present application are shown, and details of the technology are not disclosed, please refer to the embodiment shown in fig. 1 of the present application.
Referring to fig. 3, fig. 3 is a schematic structural diagram of a depression detection apparatus combining cerebral neuron spikes with local field potentials according to an embodiment of the present application. As shown in fig. 3, the apparatus includes:
the acquiring module 301 is configured to acquire electroencephalogram data of a target object in a depressed state and a non-depressed state, respectively, where the electroencephalogram data includes brain neuron spike data and local field potential data;
a fusion module 302, configured to perform feature fusion on the cerebral neuron spike potential data and the local field potential data corresponding to the same electroencephalogram data to obtain fused electroencephalogram data, and use the fused electroencephalogram data as training data;
an establishing module 303, configured to establish an ensemble learning network, train the ensemble learning network based on the training data, determine an optimal network parameter, and optimize the ensemble learning network based on the optimal network parameter;
the prediction module 304 is configured to perform classification prediction on the current electroencephalogram data of the target object based on the ensemble learning network to obtain at least one output classification prediction result, perform voting selection on each output classification prediction result to obtain a final classification result, and determine the depression state of the target object based on the final classification result.
In one possible implementation, the fusion module 302 includes:
the extraction unit is used for extracting the issuing number information of the cerebral neuron spike potential data, extracting the power spectrum information of the local field potential data and then respectively normalizing the issuing number information and the power spectrum information;
and the fusion unit is used for performing characteristic fusion on the distribution number information and the power spectrum information corresponding to the same electroencephalogram data through data splicing to obtain fused electroencephalogram data.
In one possible embodiment, the extraction unit comprises:
the first acquisition element is used for acquiring the total issuing number of the neurons in a preset time length from the brain neuron spike potential data through a sliding window, wherein the total issuing number is issuing number information;
and the second acquisition element is used for acquiring the spectral density within a preset time length in the local field potential data through the sliding window, wherein the spectral density is power spectrum information.
In one possible implementation, the fusion module 302 further includes:
the dividing unit is used for randomly dividing the fusion electroencephalogram data into training data and test data;
the device further comprises:
a verification module to verify the ensemble learning network based on the test data.
In one possible implementation, the establishing module 303 includes:
the system comprises an establishing unit, a learning integration unit and a learning integration unit, wherein the establishing unit is used for establishing the learning integration network, and a classifier in the learning integration network comprises a support vector machine, a k-nearest neighbor cluster and a width learning neural network;
and the optimization unit is used for training the ensemble learning network based on the training data, respectively determining optimal network parameters corresponding to the support vector machine, the k-nearest neighbor clustering and the width learning neural network, and optimizing the ensemble learning network based on each optimal network parameter.
In one possible implementation, the prediction module 304 includes:
the result calculation unit is used for performing classification prediction calculation on the current electroencephalogram data of the target object based on each classifier of the ensemble learning network to obtain three output classification prediction results;
the voting unit is used for voting and selecting each output classification prediction result to obtain a final classification result;
a state determination unit for determining a depression state of the target subject based on the final classification result.
In one embodiment, the method further comprises:
the mapping construction module is used for constructing a mapping relation between the target object and the integrated learning network;
and the searching module is used for searching the ensemble learning network corresponding to the target object based on the mapping relation when the new target object is determined.
It is clear to a person skilled in the art that the solution according to the embodiments of the present application can be implemented by means of software and/or hardware. The "unit" and "module" in this specification refer to software and/or hardware that can perform a specific function independently or in cooperation with other components, where the hardware may be, for example, a Field-Programmable Gate Array (FPGA), an Integrated Circuit (IC), or the like.
Each processing unit and/or module in the embodiments of the present application may be implemented by an analog circuit that implements the functions described in the embodiments of the present application, or may be implemented by software that executes the functions described in the embodiments of the present application.
Referring to fig. 4, a schematic structural diagram of an electronic device according to an embodiment of the present application is shown, where the electronic device may be used to implement the method in the embodiment shown in fig. 1. As shown in fig. 4, the electronic device 400 may include: at least one central processor 401, at least one network interface 404, a user interface 403, a memory 405, at least one communication bus 402.
Wherein a communication bus 402 is used to enable connective communication between these components.
The user interface 403 may include a Display screen (Display) and a Camera (Camera), and the optional user interface 403 may also include a standard wired interface and a wireless interface.
The network interface 404 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface), among others.
The central processing unit 401 may include one or more processing cores. The central processor 401 connects various parts within the entire electronic device 400 using various interfaces and lines, and performs various functions of the terminal 400 and processes data by executing or executing instructions, programs, code sets, or instruction sets stored in the memory 405 and calling data stored in the memory 405. Alternatively, the central Processing unit 401 may be implemented in at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), and Programmable Logic Array (PLA). The Central Processing Unit 401 may integrate one or a combination of a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a modem, and the like. Wherein, the CPU mainly processes an operating system, a user interface, an application program and the like; the GPU is used for rendering and drawing the content required to be displayed by the display screen; the modem is used to handle wireless communications. It is to be understood that the modem may be implemented by a single chip without being integrated into the central processor 401.
The Memory 405 may include a Random Access Memory (RAM) or a Read-Only Memory (Read-Only Memory). Optionally, the memory 405 includes a non-transitory computer-readable medium. The memory 405 may be used to store instructions, programs, code sets, or instruction sets. The memory 405 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing the various method embodiments described above, and the like; the storage data area may store data and the like referred to in the above respective method embodiments. The memory 405 may alternatively be at least one memory device located remotely from the central processor 401 as previously described. As shown in fig. 4, memory 405, which is a type of computer storage medium, may include an operating system, a network communication module, a user interface module, and program instructions.
In the electronic device 400 shown in fig. 4, the user interface 403 is mainly used as an interface for providing input for a user, and acquiring data input by the user; and central processor 401 may be used to invoke a depression detection application that combines cerebral neuron spikes with local field potentials stored in memory 405, and specifically performs the following operations:
acquiring electroencephalogram data of a target object in a depressed state and a non-depressed state respectively, wherein the electroencephalogram data comprise cerebral neuron spike potential data and local field potential data;
performing feature fusion on the brain neuron spike potential data and the local field potential data corresponding to the same electroencephalogram data to obtain fused electroencephalogram data, and taking the fused electroencephalogram data as training data;
establishing an ensemble learning network, training the ensemble learning network based on the training data, determining optimal network parameters, and optimizing the ensemble learning network based on the optimal network parameters;
classifying and predicting the current electroencephalogram data of the target object based on the ensemble learning network to obtain at least one output classification prediction result, voting and selecting each output classification prediction result to obtain a final classification result, and determining the depression state of the target object based on the final classification result.
The present application also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the above-described method. The computer-readable storage medium may include, but is not limited to, any type of disk including floppy disks, optical disks, DVD, CD-ROMs, microdrive, and magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic or optical cards, nanosystems (including molecular memory ICs), or any type of media or device suitable for storing instructions and/or data.
It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present application is not limited by the order of acts described, as some steps may occur in other orders or concurrently depending on the application. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required in this application.
In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus may be implemented in other manners. For example, the above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one type of division of logical functions, and there may be other divisions when actually implementing, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of some service interfaces, devices or units, and may be an electrical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable memory. Based on such understanding, the technical solution of the present application may be substantially implemented or a part of or all or part of the technical solution contributing to the prior art may be embodied in the form of a software product stored in a memory, and including several 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 described in the embodiments of the present application. And the aforementioned memory comprises: various media capable of storing program codes, such as a usb disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic disk, or an optical disk.
Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by a program, which is stored in a computer-readable memory, and the memory may include: flash disks, Read-Only memories (ROMs), Random Access Memories (RAMs), magnetic or optical disks, and the like.
The above description is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure should not be limited thereby. That is, all equivalent changes and modifications made in accordance with the teachings of the present disclosure are intended to be included within the scope of the present disclosure. Embodiments of the present disclosure will be readily apparent to those skilled in the art from consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
Claims (10)
1. A method for detecting depression with binding of brain neuron spikes to local field potentials, the method comprising:
acquiring electroencephalogram data of a target object in a depressed state and a non-depressed state respectively, wherein the electroencephalogram data comprise cerebral neuron spike potential data and local field potential data;
performing feature fusion on the brain neuron spike potential data and the local field potential data corresponding to the same electroencephalogram data to obtain fused electroencephalogram data, and taking the fused electroencephalogram data as training data;
establishing an ensemble learning network, training the ensemble learning network based on the training data, determining optimal network parameters, and optimizing the ensemble learning network based on the optimal network parameters;
classifying and predicting the current electroencephalogram data of the target object based on the ensemble learning network to obtain at least one output classification prediction result, voting and selecting each output classification prediction result to obtain a final classification result, and determining the depression state of the target object based on the final classification result.
2. The method of claim 1, wherein said performing feature fusion on said brain neuron spike data and local field potential data corresponding to the same electroencephalogram data to obtain fused electroencephalogram data comprises:
extracting the information of the number of issues of the cerebral neuron spike potential data, extracting the power spectrum information of the local field potential data, and then respectively normalizing the information of the number of issues and the power spectrum information;
and performing characteristic fusion on the release number information and the power spectrum information corresponding to the same electroencephalogram data through data splicing to obtain fused electroencephalogram data.
3. The method of claim 2, wherein extracting information on the number of firings of the cerebral neuron spike data and extracting information on the power spectrum of the local field potential data comprises:
acquiring the total issuing number of neurons in preset time duration from the brain neuron spike potential data through a sliding window, wherein the total issuing number is issuing number information;
and acquiring the frequency spectrum density in a preset time length in the local field potential data through the sliding window, wherein the frequency spectrum density is power spectrum information.
4. The method of claim 1, wherein said using said fused brain electrical data as training data comprises:
dividing the fused electroencephalogram data into training data and testing data;
after optimizing the ensemble learning network based on the optimal network parameters, the method further includes:
validating the ensemble learning network based on the test data.
5. The method of claim 1, wherein the establishing an ensemble learning network, training the ensemble learning network based on the training data, determining optimal network parameters, and optimizing the ensemble learning network based on the optimal network parameters comprises:
establishing an ensemble learning network, wherein a classifier in the ensemble learning network comprises a support vector machine, a k-nearest neighbor clustering and a width learning neural network;
training the ensemble learning network based on the training data, respectively determining optimal network parameters corresponding to the support vector machine, the k-nearest neighbor clustering and the width learning neural network, and optimizing the ensemble learning network based on each optimal network parameter.
6. The method of claim 5, wherein the classifying and predicting the current brain electrical data of the target object based on the ensemble learning network to obtain at least one output classification prediction result, voting and selecting each output classification prediction result to obtain a final classification result, and determining the depression state of the target object based on the final classification result comprises:
based on each classifier of the ensemble learning network, performing classification prediction calculation on the current electroencephalogram data of the target object respectively to obtain three output classification prediction results;
voting and selecting each output classification prediction result to obtain a final classification result;
determining a depression status of the target subject based on the final classification result.
7. The method of claim 1, further comprising:
constructing a mapping relation between a target object and the integrated learning network;
and when the new target object is determined, searching the ensemble learning network corresponding to the target object based on the mapping relation.
8. A depression detection apparatus for binding of brain neuron spikes to local field potentials, the apparatus comprising:
the acquisition module is used for respectively acquiring electroencephalogram data of a target object in a depression state and a non-depression state, wherein the electroencephalogram data comprise cerebral neuron spike potential data and local field potential data;
the fusion module is used for performing feature fusion on the brain neuron spike potential data and the local field potential data corresponding to the same electroencephalogram data to obtain fused electroencephalogram data, and taking the fused electroencephalogram data as training data;
the building module is used for building an ensemble learning network, training the ensemble learning network based on the training data, determining an optimal network parameter, and optimizing the ensemble learning network based on the optimal network parameter;
the prediction module is used for carrying out classification prediction on the current electroencephalogram data of the target object based on the ensemble learning network to obtain at least one output classification prediction result, voting and selecting each output classification prediction result to obtain a final classification result, and determining the depression state of the target object based on the final classification result.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method according to any of claims 1-7 are implemented when the computer program is executed by the processor.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210117132.7A CN114403899B (en) | 2022-02-08 | 2022-02-08 | Depression detection device combining brain neuron spike potential and local field potential |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210117132.7A CN114403899B (en) | 2022-02-08 | 2022-02-08 | Depression detection device combining brain neuron spike potential and local field potential |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114403899A true CN114403899A (en) | 2022-04-29 |
| CN114403899B CN114403899B (en) | 2023-07-25 |
Family
ID=81278339
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210117132.7A Active CN114403899B (en) | 2022-02-08 | 2022-02-08 | Depression detection device combining brain neuron spike potential and local field potential |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114403899B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114925734A (en) * | 2022-07-20 | 2022-08-19 | 浙江大学 | Online neuron classification method based on neural mimicry calculation |
| CN115381467A (en) * | 2022-10-31 | 2022-11-25 | 浙江浙大西投脑机智能科技有限公司 | Attention mechanism-based time-frequency information dynamic fusion decoding method and device |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106983505A (en) * | 2017-05-08 | 2017-07-28 | 天津医科大学 | A kind of neuroelectricity activity dependence analysis method based on comentropy |
| US20170251943A1 (en) * | 2014-08-26 | 2017-09-07 | Per Fredrik PETERSSON | Systems Level State Characteristics in Experimental Treatment of Disease |
| CN108961219A (en) * | 2018-06-15 | 2018-12-07 | 清华大学 | Natural image method for reconstructing based on local field potentials amplitude phase compound characteristics |
| KR20200018868A (en) * | 2018-08-13 | 2020-02-21 | 한국과학기술원 | Method for Adaptive EEG signal processing using reinforcement learning and System Using the same |
| CN110876626A (en) * | 2019-11-22 | 2020-03-13 | 兰州大学 | Depression detection system based on optimal lead selection of multi-lead electroencephalogram |
| CN111568446A (en) * | 2020-05-28 | 2020-08-25 | 兰州大学 | Portable electroencephalogram depression detection system combined with demographic attention mechanism |
| CN112568913A (en) * | 2020-12-23 | 2021-03-30 | 中国人民解放军总医院第四医学中心 | Electroencephalogram signal acquisition device and method |
| CN112699960A (en) * | 2021-01-11 | 2021-04-23 | 华侨大学 | Semi-supervised classification method and equipment based on deep learning and storage medium |
| CN113052113A (en) * | 2021-04-02 | 2021-06-29 | 中山大学 | Depression identification method and system based on compact convolutional neural network |
| CN113180669A (en) * | 2021-05-12 | 2021-07-30 | 中国人民解放军中部战区总医院 | Emotional regulation training system and method based on nerve feedback technology |
| CN113378737A (en) * | 2021-06-18 | 2021-09-10 | 河北大学 | Implanted brain-computer interface neuron spike potential classification method |
| CN113397563A (en) * | 2021-07-22 | 2021-09-17 | 北京脑陆科技有限公司 | Training method, device, terminal and medium for depression classification model |
| CN113812958A (en) * | 2021-09-22 | 2021-12-21 | 杭州诺为医疗技术有限公司 | Brain internal stimulation and detection system and method |
| CN113951898A (en) * | 2021-10-15 | 2022-01-21 | 浙江大学 | P300 electroencephalogram signal detection method and device for data migration, electronic device and medium |
-
2022
- 2022-02-08 CN CN202210117132.7A patent/CN114403899B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170251943A1 (en) * | 2014-08-26 | 2017-09-07 | Per Fredrik PETERSSON | Systems Level State Characteristics in Experimental Treatment of Disease |
| CN106983505A (en) * | 2017-05-08 | 2017-07-28 | 天津医科大学 | A kind of neuroelectricity activity dependence analysis method based on comentropy |
| CN108961219A (en) * | 2018-06-15 | 2018-12-07 | 清华大学 | Natural image method for reconstructing based on local field potentials amplitude phase compound characteristics |
| KR20200018868A (en) * | 2018-08-13 | 2020-02-21 | 한국과학기술원 | Method for Adaptive EEG signal processing using reinforcement learning and System Using the same |
| CN110876626A (en) * | 2019-11-22 | 2020-03-13 | 兰州大学 | Depression detection system based on optimal lead selection of multi-lead electroencephalogram |
| CN111568446A (en) * | 2020-05-28 | 2020-08-25 | 兰州大学 | Portable electroencephalogram depression detection system combined with demographic attention mechanism |
| CN112568913A (en) * | 2020-12-23 | 2021-03-30 | 中国人民解放军总医院第四医学中心 | Electroencephalogram signal acquisition device and method |
| CN112699960A (en) * | 2021-01-11 | 2021-04-23 | 华侨大学 | Semi-supervised classification method and equipment based on deep learning and storage medium |
| CN113052113A (en) * | 2021-04-02 | 2021-06-29 | 中山大学 | Depression identification method and system based on compact convolutional neural network |
| CN113180669A (en) * | 2021-05-12 | 2021-07-30 | 中国人民解放军中部战区总医院 | Emotional regulation training system and method based on nerve feedback technology |
| CN113378737A (en) * | 2021-06-18 | 2021-09-10 | 河北大学 | Implanted brain-computer interface neuron spike potential classification method |
| CN113397563A (en) * | 2021-07-22 | 2021-09-17 | 北京脑陆科技有限公司 | Training method, device, terminal and medium for depression classification model |
| CN113812958A (en) * | 2021-09-22 | 2021-12-21 | 杭州诺为医疗技术有限公司 | Brain internal stimulation and detection system and method |
| CN113951898A (en) * | 2021-10-15 | 2022-01-21 | 浙江大学 | P300 electroencephalogram signal detection method and device for data migration, electronic device and medium |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114925734A (en) * | 2022-07-20 | 2022-08-19 | 浙江大学 | Online neuron classification method based on neural mimicry calculation |
| CN114925734B (en) * | 2022-07-20 | 2022-11-25 | 浙江大学 | Online neuron classification method based on neural mimicry calculation |
| CN115381467A (en) * | 2022-10-31 | 2022-11-25 | 浙江浙大西投脑机智能科技有限公司 | Attention mechanism-based time-frequency information dynamic fusion decoding method and device |
| CN115381467B (en) * | 2022-10-31 | 2023-03-10 | 浙江浙大西投脑机智能科技有限公司 | Attention mechanism-based time-frequency information dynamic fusion decoding method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114403899B (en) | 2023-07-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hameed et al. | Multi-class skin diseases classification using deep convolutional neural network and support vector machine | |
| CN114403899B (en) | Depression detection device combining brain neuron spike potential and local field potential | |
| Siuly et al. | EEG signal classification based on simple random sampling technique with least square support vector machine | |
| Chén et al. | Building a machine-learning framework to remotely assess Parkinson's disease using smartphones | |
| Shaban | Automated screening of parkinson's disease using deep learning based electroencephalography | |
| CN115662576B (en) | Method and system for generating neurofeedback training paradigm of associated cognitive disorder conditions | |
| US8271414B2 (en) | Network characterization, feature extraction and application to classification | |
| CN115414041A (en) | Autism assessment device, method, terminal device and medium based on electroencephalogram data | |
| CN115668395A (en) | System and method for processing retinal signal data and identifying conditions | |
| CN113662560A (en) | Method for detecting seizure-like discharge between attacks, storage medium and device | |
| Wang et al. | Automated rest eeg-based diagnosis of depression and schizophrenia using a deep convolutional neural network | |
| Panicacci et al. | Population health management exploiting machine learning algorithms to identify high-risk patients | |
| Chen et al. | DCTNet: Hybrid deep neural network-based EEG signal for detecting depression | |
| CN114742107A (en) | Method for identifying perception signal in information service and related equipment | |
| Madruga et al. | Addressing smartphone mismatch in Parkinson’s disease detection aid systems based on speech | |
| CN113197545B (en) | Epilepsy detection system based on graph attention residual error network and focus loss | |
| CN110889836A (en) | Image data analysis method and device, terminal equipment and storage medium | |
| US11786156B2 (en) | Method and apparatus for use in detecting malingering by a first subject in tests of physical and/or mental function of the first subject | |
| KR20210119081A (en) | Method for providing information of major depressive disorders and device for providing information of major depressive disorders using the same | |
| CN114305423B (en) | Depression state indication determining device based on neuron spike signal | |
| Srikanth | Parkinson Disease Detection Using Various Machine Learning Algorithms | |
| Singh et al. | Parkinson’s disease detection using machine learning | |
| Li et al. | A smart detection technology for personal ECG monitoring via chaos-based data mapping strategy | |
| Silaen et al. | EEG Signal Processing For Motor Imagery Direction of Hand Movement Using the Brain Computer Interface | |
| Jain et al. | An efficient feature extraction technique and novel normalization method to improve EMG signal classification |
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 |