CN112568868B - Automatic quantitative analysis method and device for electrophysiological signals of epilepsy model - Google Patents
Automatic quantitative analysis method and device for electrophysiological signals of epilepsy model Download PDFInfo
- Publication number
- CN112568868B CN112568868B CN202011107133.0A CN202011107133A CN112568868B CN 112568868 B CN112568868 B CN 112568868B CN 202011107133 A CN202011107133 A CN 202011107133A CN 112568868 B CN112568868 B CN 112568868B
- Authority
- CN
- China
- Prior art keywords
- data
- spike
- neural network
- mean value
- duration
- 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
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/40—Detecting, measuring or recording for evaluating the nervous system
- A61B5/4076—Diagnosing or monitoring particular conditions of the nervous system
- A61B5/4094—Diagnosing or monitoring seizure diseases, e.g. epilepsy
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4848—Monitoring or testing the effects of treatment, e.g. of medication
-
- 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
-
- 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
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7282—Event detection, e.g. detecting unique waveforms indicative of a medical condition
-
- 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
-
- 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B2503/00—Evaluating a particular growth phase or type of persons or animals
- A61B2503/40—Animals
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B2503/00—Evaluating a particular growth phase or type of persons or animals
- A61B2503/42—Evaluating a particular growth phase or type of persons or animals for laboratory research
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- Theoretical Computer Science (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Pathology (AREA)
- Veterinary Medicine (AREA)
- Artificial Intelligence (AREA)
- Physiology (AREA)
- Neurology (AREA)
- Neurosurgery (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention provides an automatic quantitative analysis method for an epilepsia model electrophysiological signal, which is characterized in that a characteristic value of data recorded by an epilepsia model neuroelectrophysiology is used as input of a neural network, a first interval number N and a first interval duration T obtained by a sliding window and a box separation algorithm are used as output of the neural network, and the number and duration of epilepsia occurrence are calculated by the neural network algorithm to replace manual judgment. The method can quickly and accurately evaluate the electrophysiological signals of the epilepsy model, is basically consistent with the result of artificial evaluation, and can provide quick and accurate efficacy evaluation for screening high-throughput antiepileptic drugs.
Description
Technical Field
The invention relates to the field of analysis of epilepsy electrophysiological signals, in particular to an automatic identification and quantitative analysis algorithm for electrophysiological signals of an in-vitro model of epilepsy.
Background
At present, in-vitro and in-vivo models of epilepsy based on brain slices and zebra fishes are widely used for researching epileptic pathological mechanisms and screening antiepileptic drugs, and electrophysiological signals are the gold standard for evaluating epileptogenesis and drug intervention effectiveness.
High-throughput drug screening relates to massive electrophysiological data processing and analysis, and manual analysis requires that an analyst have enough epilepsy electrophysiological signal identification experience and manually measure the occurrence frequency and the occurrence duration of an epileptic event, which is time-consuming and labor-consuming, and individual deviation is easily generated according to the epilepsy signal waveform identified by human eyes, and the final result is not completely objective.
Disclosure of Invention
The invention aims to provide an automatic quantitative analysis method for electrophysiological signals of an epilepsy model, which is used for solving the technical problem of dependence of manual analysis on experience and manpower.
In order to achieve the above purpose, the invention provides the following technical scheme:
aiming at an automatic quantitative analysis method of an electrophysiological signal of an epilepsy model,
obtaining a training set, wherein the training set is parameters of data recorded by neuroelectrophysiology of an epilepsy model, and the parameters comprise total duration, signal-to-noise ratio, background signal mean value, sampling frequency, spike/spike absolute height mean value, spike/spike width mean value, spike/spike first-order difference mean value, spike/spike slope mean value, first interval number N and first interval duration T;
constructing a neural network, wherein the input end of the neural network respectively represents the total time length, the signal-to-noise ratio, the background signal mean value sampling frequency, the spike/spike absolute height mean value, the spike/spike width mean value, the spike/spike first-order difference mean value and the spike/spike slope mean value; the output end of the neural network is the number N of the first intervals and the duration T of the first intervals;
the first interval number N and the duration T of the first interval are obtained by the following method:
obtaining first data, wherein the first data is data of an epilepsy model neuroelectrophysiology record;
removing high and low frequency noise, and calibrating the baseline to obtain second data;
presetting a first threshold reference, comparing the second data with the first threshold reference, and converting the second data into binary third data;
and performing sliding window processing on the third data, and performing window sliding according to the box separation distance until all the third data are subjected to sliding processing according to the following processing modes: with the sliding of the window, performing binning processing on third data in the current window every time to obtain a first binning value in the current window as fourth data;
comparing the fourth data with a preset threshold value D, wherein the preset threshold value D is a binning threshold value, selecting continuous intervals which are larger than or equal to D in the fourth data as first intervals, the number of the first intervals is N, and the duration of each first interval is first time T;
and analyzing the actually measured electrophysiological signals of the epilepsy model by using the constructed neural network to obtain the corresponding first interval number N and first interval duration T.
Further, in the invention, after the neural network outputs, the artificial judgment is carried out to judge whether the neural network receives the training data, if the neural network receives the training data, the result is output, if the neural network does not receive the training data, the number N of the first intervals and the first time T are given manually, new training data are formed, meanwhile, the preset threshold value D is modified to update the training data, the training is carried out, and the neural network is optimized to retrain the neural network until the output first interval time T of the neural network is less than or equal to 3% compared with the artificial measurement.
Further, in the invention, the first threshold reference is between 0.15mv and 0.25 mv.
Further, in the present invention, the preset threshold D is between 0.2 and 0.4.
Further, in the present invention, the sliding window duration is 0.5s, and the binning distance is 0.01s.
Further, in the present invention, during the neural network training, the ratio of the training set, the test set and the development set is 4; and (4) performing dominant screening in the N neural networks by using the test set, evaluating the prediction effect of the dominant neural network by using the development set, and finally determining the first neural network.
Further, in the invention, during the training of the neural network, the adopted activation function is a ReLU function, the adopted loss function is a mean square error MES, and the adopted advantage screening method is gradient descent including regularization.
Has the advantages that:
according to the technical scheme, the automatic analysis method for the electrophysiological signals of the epilepsy model is provided, the method is combined with a neural network, a sliding window and a data binning algorithm, the electrophysiological signals of the epilepsy model can be rapidly and accurately evaluated, results of manual evaluation are basically consistent, and rapid and accurate drug effect evaluation can be provided for high-throughput antiepileptic drug screening.
The method further optimizes the neural network according to the actual measurement result, so that the neural network is more intelligent, and the accuracy is improved.
It should be understood that all combinations of the foregoing concepts and additional concepts described in greater detail below can be considered as part of the inventive subject matter of this disclosure unless such concepts are mutually inconsistent.
The foregoing and other aspects, embodiments and features of the present teachings can be more fully understood from the following description taken in conjunction with the accompanying drawings. Additional aspects of the present invention, such as features and/or advantages of exemplary embodiments, will be apparent from the description which follows, or may be learned by practice of specific embodiments in accordance with the teachings of the present invention.
Drawings
The drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component that is illustrated in various figures may be represented by a like numeral. For purposes of clarity, not every component may be labeled in every drawing. Embodiments of various aspects of the present invention will now be described, by way of example, with reference to the accompanying drawings, in which:
FIG. 1 is a flow chart of an automated analysis system of electrophysiological signals for epilepsy models of the present invention;
FIG. 2 is a flow chart of the sliding window and binning process of the present invention;
FIG. 3 is a comparison of manual and algorithmic measurements of event duration.
Detailed Description
In order to better understand the technical content of the present invention, specific embodiments are described below with reference to the accompanying drawings.
In this disclosure, aspects of the present invention are described with reference to the accompanying drawings, in which a number of illustrative embodiments are shown. Embodiments of the present disclosure are not necessarily intended to include all aspects of the invention. It should be appreciated that the various concepts and embodiments described above, as well as those described in greater detail below, may be implemented in any of numerous ways, as the disclosed concepts and embodiments are not limited to any one implementation. In addition, some aspects of the present disclosure may be used alone, or in any suitable combination with other aspects of the present disclosure.
Summary of the application
High-throughput drug screening relates to massive electrophysiological data processing and analysis, and manual analysis requires that an analyst have enough epilepsy electrophysiological signal identification experience and manually measure the occurrence frequency and the occurrence duration of epilepsy time, which is time-consuming and labor-consuming, and individual deviation is easily generated according to the epilepsy signal waveform identified by human eyes, and the final result is not completely objective.
In view of the above technical problems, the technical solution provided by the present application has the following general idea:
the automatic quantitative analysis method for the electrophysiological signals of the epilepsy model comprises the following steps:
s101, obtaining a training set, wherein the training set is parameters of data recorded by neuroelectrophysiology of the epilepsy model and comprises total duration, signal-to-noise ratio, background signal mean value, sampling frequency, spike/spike absolute height mean value, spike/spike width mean value, spike/spike first-order difference mean value, spike/spike slope mean value, first interval number N and first interval duration T;
s102, constructing a neural network, wherein the input end of the neural network respectively represents the total time length, the signal-to-noise ratio, the background signal mean value sampling frequency, the spike/spike absolute height mean value, the spike/spike width mean value, the spike/spike first-order difference mean value and the spike/spike slope mean value; the output end of the neural network is the number N of the first intervals and the duration T of the first intervals;
the first interval number N and the first interval duration T are obtained by the following method:
s1021, obtaining first data, wherein the first data are data recorded by the neuroelectrophysiology of the epilepsy model;
s1022, removing high and low frequency noises, and calibrating the baseline to obtain second data;
s1023, presetting a first threshold reference, comparing the second data with the first threshold reference, and converting the second data into binary third data;
s1024, performing sliding window processing on the third data, and performing window sliding according to the box separation distance until all the third data are subjected to sliding processing according to the following processing modes: with the sliding of the window, performing binning processing on third data in the current window every time to obtain a first binning value in the current window as fourth data;
s1025, comparing the fourth data with a preset threshold value D, wherein the preset threshold value D is a binning threshold value, selecting the continuous intervals which are larger than or equal to D in the fourth data as first intervals, the number of the first intervals is N, and the duration of each first interval is first time T;
s103, analyzing the measured electrophysiological signals of the epilepsy model by using the constructed neural network to obtain the corresponding first interval number N and first interval duration T.
Having described the principles of the present application, various non-limiting embodiments thereof will now be described in detail with reference to the accompanying drawings.
The first embodiment,
The scheme integrates a sliding window and a box separation technology, processes the time-series brain wave signals, and obtains reliable first interval quantity N and first interval duration T for forming training data of the neural network.
And then, a plurality of characteristics related to epilepsy response in the brain wave signals are used as input in neural network training data, and the training data, the number N of the first intervals and the duration T of the first intervals form training data together to complete the training of the neural network, so that the neural network can prepare for processing the signals.
In the above process, in the process of obtaining the training data, a high-pass filter and a low-pass filter are used for denoising, specifically, in the embodiment of the present invention, the high-pass filter is set to 0.1Hz, the low-pass filter is set to 5000 Hz, and the notch filter is set to 50-60Hz.
The first data is time series data, so that the baseline drift is corrected by adopting a moving average method during baseline correction.
The first threshold reference and the binning threshold are both set according to the signal-to-noise ratio. Specifically, the first threshold value reference is between 0.15mv and 0.25mv, and the preset threshold value D is between 0.2 and 0.4. And binarizing the signal by judging whether the signal exceeds the first threshold reference, setting the data points exceeding the first threshold reference as 1, and setting the data points not exceeding the first threshold reference as 0, and finally forming a binary character string. According to a binning threshold, identifying the starting and ending time point of each epileptic event, taking the point when the binning value reaches or exceeds the binning threshold as a starting time point T1, taking the point when the binning value begins to be lower than the binning threshold as an ending time point T2, recording the number of the events and calculating the duration of the events: t = T2-T1.
Further, the precision of the processing method is adjusted through the time length of the sliding window and the time length of the box separation, in the embodiment of the invention, the time length of the sliding window is set to be 0.5s, and the time length of the box separation is set to be 0.01s.
Training a neural network by using the data as training data, wherein the neural network comprises a training set and a test set and a development set in addition to the training set during training, the proportion of the training set, the test set and the development set is 4; and (4) performing dominant screening in the N neural networks by using the test set, evaluating the prediction effect of the dominant neural network by using the development set, and finally determining the first neural network. During training of the neural network, an adopted activation function is a ReLU function, an adopted loss function is a mean square error MES, and an adopted advantage screening method is gradient descent including regularization.
The trained neural network can identify typical epileptic brain waves and output corresponding first interval number N and first interval duration T.
After the neural network is output, manual verification is used for improving the reliability of data and further optimizing the neural network. During manual verification, manually judging whether the received data are received or not, outputting a result if the received data are received, and manually giving the first interval number N and the first time T if the received data are not received, and forming new training data to retrain the neural network until the no-difference of the neural network is less than or equal to 3% compared with the manual measurement.
During manual verification, data with the time length of 10 s-20 s are randomly taken, the event time length is manually measured, and the measured time length is compared with the measurement result given by the algorithm;
1) If the time length deviation of the events is less than or equal to 3%, the data output by the system is received;
2) If the event duration algorithm result is higher than the artificial result by more than 3%, the error of the data output by the system is large, the result given by the artificial is used as training data, the training data is updated again by adjusting the height and the box dividing threshold value to be 0.05, and then training is carried out, so that the neural network is optimized;
3) If the result of the event duration algorithm is lower than the artificial result and exceeds 3%, the error of the data output by the system is large, the result given by the artificial is used as training data, the training data is updated again by adjusting the lower binning threshold value to be 0.05, and then training is carried out, so that the neural network is optimized.
As shown in fig. 3, the ratio of the measurement result output by the optimized neural network of the present invention to the artificial measurement result falls around the x = y straight line, and it can be seen that the event duration of the algorithm measurement and the artificial measurement is substantially consistent.
Example II,
An automatic quantitative analysis device for electrophysiological signals of epilepsy model comprises
The first obtaining module is used for obtaining a training set, wherein the training set is parameters of data recorded by neuroelectrophysiology of the epilepsy model and comprises total duration, a signal-to-noise ratio, a background signal mean value, a sampling frequency, a spike/spike absolute height mean value, a spike/spike width mean value, a spike/spike first-order difference mean value, a spike/spike slope mean value, a first interval number N and a first interval duration T;
the first construction module is used for constructing a neural network, and the input end of the neural network respectively represents total time length, signal-to-noise ratio, background signal mean value, sampling frequency, spike/spike absolute height mean value, spike/spike width mean value, spike/spike first-order difference mean value and spike/spike slope mean value; the output end of the neural network is the number N of the first intervals and the duration T of the first intervals;
the first interval number N and the first interval duration T are obtained by the following method:
obtaining first data, wherein the first data is data of an epilepsy model neuroelectrophysiology record;
removing high and low frequency noise, and calibrating the baseline to obtain second data;
presetting a first threshold reference, comparing the second data with the first threshold reference, and converting the second data into binaryzation third data;
and performing sliding window processing on the third data, and performing window sliding according to the box separation distance until all the third data are subjected to sliding processing according to the following processing modes: with the sliding of the window, performing binning processing on third data in the current window every time to obtain a first binning value in the current window as fourth data;
comparing the fourth data with a preset threshold value D, wherein the preset threshold value D is a binning threshold value, selecting the continuous intervals which are greater than or equal to D in the fourth data as first intervals, the number of the first intervals is N, and the duration of each first interval is first time T;
and analyzing the measured electrophysiological signals of the epilepsy model by using the constructed neural network through the first output module to obtain the corresponding first interval number N and the first interval duration T.
Example III,
Based on the same inventive concept as the method for the automatic quantitative analysis of the epilepsy model electrophysiological signal in the foregoing embodiments, the present invention further provides an exemplary electronic device, an apparatus for the automatic quantitative analysis of the epilepsy model electrophysiological signal, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method for the automatic quantitative analysis of the epilepsy model electrophysiological signal when executing the program.
One or more technical solutions in the embodiments of the present application at least have one or more of the following technical effects: the method can quickly and accurately evaluate the electrophysiological signals of the epilepsy model, is basically consistent with the result of artificial evaluation, and can provide quick and accurate efficacy evaluation for screening high-throughput antiepileptic drugs.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Although the present invention has been described with reference to the preferred embodiments, it is not intended to be limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention. Therefore, the protection scope of the present invention should be determined by the appended claims.
Claims (8)
1. The automatic quantitative analysis method for the electrophysiological signals of the epilepsy model is used for screening the medicines, and is characterized in that:
obtaining a training set, wherein the training set is parameters of data recorded by neuroelectrophysiology of an epilepsy model, and the parameters comprise total duration, signal-to-noise ratio, background signal mean value, sampling frequency, spike/spike absolute height mean value, spike/spike width mean value, spike/spike first-order difference mean value, spike/spike slope mean value, first interval number N and first interval duration T;
constructing a neural network, wherein the input end of the neural network respectively represents the total time length, the signal-to-noise ratio, the background signal mean value sampling frequency, the spike/spike absolute height mean value, the spike/spike width mean value, the spike/spike first-order difference mean value and the spike/spike slope mean value; the output end of the neural network is the number N of the first intervals and the duration T of the first intervals;
the first interval number N and the first interval duration T are obtained by the following method:
obtaining first data, wherein the first data is data of an epilepsy model neuroelectrophysiology record;
removing high and low frequency noise, and calibrating the baseline to obtain second data;
presetting a first threshold reference, comparing the second data with the first threshold reference, and converting the second data into binary third data;
and performing sliding window processing on the third data, and performing window sliding according to the box separation distance until all the third data are subjected to sliding processing according to the following processing modes: with the sliding of the window, performing binning processing on third data in the current window every time to obtain a first binning value in the current window as fourth data;
comparing the fourth data with a preset threshold value D, wherein the preset threshold value D is a binning threshold value, selecting continuous intervals which are larger than or equal to D in the fourth data as first intervals, the number of the first intervals is N, and the duration of each first interval is first time T;
analyzing the measured electrophysiological signals of the epilepsy model by using the constructed neural network to obtain the corresponding first interval number N and first interval duration T;
after the neural network is output, manually judging whether the neural network is received or not, if the neural network is received, outputting a result, if the neural network is not received, manually giving the first interval number N and the first time T, forming new training data, modifying a preset threshold value D, updating the training data, training again, and optimizing the neural network to retrain the neural network until the output first interval duration T of the neural network is less than or equal to 3% compared with the manual measurement.
2. The method for drug screening based on the automated quantitative analysis of electrophysiological signals of epilepsy models according to claim 1, wherein the method comprises the steps of: the first threshold value reference is between 0.15mv and 0.25 mv.
3. The method for automatically and quantitatively analyzing electrophysiological signals of epilepsy models according to claim 1, wherein the method comprises: the preset threshold value D is between 0.2 and 0.4.
4. The method for automated quantitative analysis of electrophysiological signals for epilepsy modeling for drug screening according to any one of claims 1 to 3, wherein the method comprises the steps of: the time length of the sliding window is 0.5s, and the box separation distance is 0.01s.
5. The method for automatically and quantitatively analyzing electrophysiological signals of epilepsy models according to claim 4, wherein the method comprises: during the neural network training, the proportion of a training set, a test set and a development set is 4; and (4) performing dominant screening in the N neural networks by using the test set, evaluating the prediction effect of the dominant neural network by using the development set, and finally determining the first neural network.
6. The method for automatically and quantitatively analyzing electrophysiological signals of epilepsy models for drug screening according to claim 5, wherein the method comprises: during training of the neural network, an adopted activation function is a ReLU function, an adopted loss function is a mean square error MES, and an adopted advantage screening method is gradient descent including regularization.
7. Medicine screening is with automatic quantitative analysis device to epileptic model electrophysiological signal, its characterized in that: comprises that
The first obtaining module is used for obtaining a training set, wherein the training set is parameters of data recorded by neuroelectrophysiology of the epilepsy model and comprises total duration, a signal-to-noise ratio, a background signal mean value, a sampling frequency, a spike/spike absolute height mean value, a spike/spike width mean value, a spike/spike first-order difference mean value, a spike/spike slope mean value, a first interval number N and a first interval duration T;
the first construction module is used for constructing a neural network, and the input end of the neural network respectively represents total time length, signal-to-noise ratio, background signal mean value, sampling frequency, spike/spike absolute height mean value, spike/spike width mean value, spike/spike first-order difference mean value and spike/spike slope mean value; the output end of the neural network is the number N of the first intervals and the duration T of the first intervals;
the first interval number N and the duration T of the first interval are obtained by the following method:
obtaining first data, wherein the first data is data of an epilepsy model neuroelectrophysiology record;
removing high and low frequency noise, and calibrating the baseline to obtain second data;
presetting a first threshold reference, comparing the second data with the first threshold reference, and converting the second data into binaryzation third data;
and performing sliding window processing on the third data, and performing window sliding according to the box separation distance until all the third data are subjected to sliding processing according to the following processing modes: with the sliding of the window, performing binning processing on third data in the current window every time to obtain a first binning numerical value in the current window as fourth data;
comparing the fourth data with a preset threshold value D, wherein the preset threshold value D is a binning threshold value, selecting continuous intervals which are larger than or equal to D in the fourth data as first intervals, the number of the first intervals is N, and the duration of each first interval is first time T;
the first output module is used for analyzing the measured electrophysiological signals of the epilepsy model by using the constructed neural network to obtain the corresponding first interval number N and the first interval duration T;
and the first optimization module is used for manually judging whether the received data is received or not after the neural network is output, outputting a result if the received data is received, manually giving the first interval number N and the first time T if the received data is not received, forming new training data, modifying a preset threshold value D, updating the training data and then training, and optimizing the neural network to retrain the neural network until the output first interval duration T of the neural network is less than or equal to 3% compared with the manual measurement.
8. Medicine screening is with automatic quantitative analysis device to epileptic model electrophysiological signal, its characterized in that: comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method according to any one of claims 1 to 6 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011107133.0A CN112568868B (en) | 2020-10-16 | 2020-10-16 | Automatic quantitative analysis method and device for electrophysiological signals of epilepsy model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011107133.0A CN112568868B (en) | 2020-10-16 | 2020-10-16 | Automatic quantitative analysis method and device for electrophysiological signals of epilepsy model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112568868A CN112568868A (en) | 2021-03-30 |
| CN112568868B true CN112568868B (en) | 2022-12-23 |
Family
ID=75119961
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011107133.0A Active CN112568868B (en) | 2020-10-16 | 2020-10-16 | Automatic quantitative analysis method and device for electrophysiological signals of epilepsy model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112568868B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113533264A (en) * | 2021-04-06 | 2021-10-22 | 浙江赛微思生物科技有限公司 | Biomarker, detection method thereof and application of biomarker in epilepsy pathological research and antiepileptic drug screening |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101991569A (en) * | 2009-08-20 | 2011-03-30 | 四川大学华西医院 | Rhesus monkey temporal lobe epilepsy kindling model and drug screening method thereof |
| WO2017060400A1 (en) * | 2015-10-07 | 2017-04-13 | Neurovive Pharmaceutical Ab | Protected carboxylic acid-based metabolites for the treatment of disesases related to mitochondrial dysfunctions |
| CN107518896A (en) * | 2017-07-12 | 2017-12-29 | 中国科学院计算技术研究所 | A kind of myoelectricity armlet wearing position Forecasting Methodology and system |
| CN110197132A (en) * | 2019-05-13 | 2019-09-03 | 径卫视觉科技（上海）有限公司 | A kind of method and corresponding system of living body faces for identification |
| CN111462921A (en) * | 2020-04-20 | 2020-07-28 | 山东大学 | Epilepsy medication recommendation method and system |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012167140A1 (en) * | 2011-06-01 | 2012-12-06 | Drexel University | System and method of detecting and predicting seizures |
| JP2016521998A (en) * | 2013-03-15 | 2016-07-28 | サイモン、アダム、ジェイ． | Multimodal drug diagnostic evaluation of brain health |
| CN108577834B (en) * | 2018-05-17 | 2019-07-02 | 西安交通大学 | A method of it is detected automatically for epilepsy interphase spike |
| US11612750B2 (en) * | 2019-03-19 | 2023-03-28 | Neuropace, Inc. | Methods and systems for optimizing therapy using stimulation mimicking natural seizures |
| CN110321783A (en) * | 2019-05-07 | 2019-10-11 | 北京慧脑云计算有限公司 | A kind of MEG spike detection method and system based on 1D convolutional neural networks |
| CN110811609B (en) * | 2019-10-28 | 2022-11-22 | 杭州电子科技大学 | Epileptic spike intelligent detection device based on self-adaptive template matching and machine learning algorithm fusion |
| CN111150393B (en) * | 2020-02-19 | 2023-03-28 | 杭州电子科技大学 | Electroencephalogram epilepsy spike discharge joint detection method based on LSTM multichannel |
| CN111419222B (en) * | 2020-03-08 | 2023-04-14 | 杭州电子科技大学 | BNI-based epileptic seizure signal detection method |
| CN111340142B (en) * | 2020-05-14 | 2020-08-14 | 南京慧脑云计算有限公司 | Epilepsia magnetoencephalogram spike automatic detection method and tracing positioning system |
-
2020
- 2020-10-16 CN CN202011107133.0A patent/CN112568868B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101991569A (en) * | 2009-08-20 | 2011-03-30 | 四川大学华西医院 | Rhesus monkey temporal lobe epilepsy kindling model and drug screening method thereof |
| WO2017060400A1 (en) * | 2015-10-07 | 2017-04-13 | Neurovive Pharmaceutical Ab | Protected carboxylic acid-based metabolites for the treatment of disesases related to mitochondrial dysfunctions |
| CN107518896A (en) * | 2017-07-12 | 2017-12-29 | 中国科学院计算技术研究所 | A kind of myoelectricity armlet wearing position Forecasting Methodology and system |
| CN110197132A (en) * | 2019-05-13 | 2019-09-03 | 径卫视觉科技（上海）有限公司 | A kind of method and corresponding system of living body faces for identification |
| CN111462921A (en) * | 2020-04-20 | 2020-07-28 | 山东大学 | Epilepsy medication recommendation method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112568868A (en) | 2021-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shiao et al. | SVM-based system for prediction of epileptic seizures from iEEG signal | |
| Birot et al. | Automatic detection of fast ripples | |
| CN112735094B (en) | Geological disaster prediction method and device based on machine learning and electronic equipment | |
| Rana et al. | Seizure detection using the phase-slope index and multichannel ECoG | |
| US11520672B2 (en) | Anomaly detection device, anomaly detection method and storage medium | |
| US9294237B2 (en) | Method for performing joint jitter and amplitude noise analysis on a real time oscilloscope | |
| JP2022536552A (en) | Seizure detection system and method based on changes in electroencephalogram (EEG) nonlinearity | |
| CN112568868B (en) | Automatic quantitative analysis method and device for electrophysiological signals of epilepsy model | |
| CN113568774A (en) | Real-time anomaly detection method for multi-dimensional time sequence data by using unsupervised deep neural network | |
| JP2023099043A (en) | Systems and methods for seizure detection based on changes in electroencephalogram (eeg) non-linearities | |
| US20160045120A1 (en) | Systems and methods for spike sorting | |
| CN114443331A (en) | Time series data abnormity detection method and device | |
| CN111768287A (en) | Period identification method, period identification device, server and readable storage medium | |
| KR20140016559A (en) | Apparatus and method for monitoring disease status based on time-series data modeling | |
| EP3304365B1 (en) | System for the detection and the early prediction of the approaching of exacerbations in patients suffering from chronic obstructive broncopneumaty | |
| CN112057068A (en) | Epilepsia pathological data classification method and device and storage medium | |
| CN114782837B (en) | Plant estimation method, plant estimation device, electronic equipment and storage medium | |
| CN115600747A (en) | Tunnel state monitoring management method and system based on Internet of things | |
| CN109102143B (en) | Yield monitoring method and device | |
| CN113349797B (en) | System for predicting epileptic seizure based on electroencephalogram critical nuclear parameters | |
| CN116448062B (en) | Bridge settlement deformation detection method, device, computer and storage medium | |
| Thieu et al. | Diagnosis of epilepsy in patients based on the classification of EEG signals using fast Fourier transform | |
| CN117454096B (en) | Motor production quality detection method and system | |
| CN116636815B (en) | Electroencephalogram signal-based sleeping quality assessment method and system for underwater operators | |
| EP3925537B1 (en) | Method and system for processing an electroencephalogram signal |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20221114 Address after: 215000, 17 / F, Tiancheng Times Business Plaza, No. 58, qinglonggang Road, high speed railway new town, Xiangcheng District, Suzhou City, Jiangsu Province Applicant after: SUZHOU SEMEK GENE TECHNOLOGY Co.,Ltd. Address before: Room 401 and 407, building A2, 555 Chuangye Road, Dayun Town, Jiashan County, Jiaxing City, Zhejiang Province, 314100 Applicant before: Zhejiang saiweisi Biotechnology Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |