CN112155523B - Pulse signal feature extraction and classification method based on modal energy principal component ratio quantification - Google Patents
Pulse signal feature extraction and classification method based on modal energy principal component ratio quantification Download PDFInfo
- Publication number
- CN112155523B CN112155523B CN202011035168.8A CN202011035168A CN112155523B CN 112155523 B CN112155523 B CN 112155523B CN 202011035168 A CN202011035168 A CN 202011035168A CN 112155523 B CN112155523 B CN 112155523B
- Authority
- CN
- China
- Prior art keywords
- modal
- signal
- component
- signals
- imf
- 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
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
-
- 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
- A61B5/4854—Diagnosis based on concepts of traditional oriental medicine
-
- 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
- A61B5/7235—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
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Abstract
The invention relates to a pulse signal feature extraction and classification method based on modal energy principal component ratio quantification, belonging to the technical field of pulse signal feature extraction and classification; the technical problem to be solved is as follows: the improvement of a pulse signal characteristic extraction and classification method based on modal energy principal component ratio quantification is provided; the technical scheme for solving the technical problem is as follows: the method comprises the following steps: acquiring a section of normal human pulse signals through a pulse sensor, carrying out CEEMDAN self-adaptive full integration empirical mode decomposition on the acquired signals, decomposing original complex signals into mode components with different time scales, further decomposing the decomposed signals based on a Mode Energy Principal Component Ratio (MEPCR), and quantizing the decomposed signals for describing the energy principal component distribution of pulses; the invention is applied to pulse signal feature extraction and classification.
Description
Technical Field
The invention discloses a pulse signal feature extraction and classification method based on modal energy principal component ratio quantization, and belongs to the technical field of pulse signal feature extraction and classification.
Background
Traditional Chinese medicine and modern medicine show that the pulse signals of human bodies contain rich health condition information of the human bodies, in recent years, many researchers are dedicated to detection and analysis of the pulse signals of the human bodies, and diagnosis methods for extracting and classifying various pulse signal characteristics are proposed and applied; the existing pulse signal processing mainly comprises the steps of carrying out characteristic quantization on signals and then classifying the signals according to characteristic vectors, the existing extraction and classification method can basically achieve the classification effect, but the classification efficiency and the effectiveness are still insufficient, and the method process needs to be improved and optimized; because the pulse signal is a typical nonlinear and non-stationary signal, the oscillation mode of the pulse signal is complex, and the problems of low accuracy and low robustness can occur when the signal is classified by adopting a single characteristic.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention aims to solve the technical problems that: the improvement of the pulse signal feature extraction and classification method based on modal energy principal component ratio quantification is provided.
In order to solve the technical problems, the invention adopts the technical scheme that: a pulse signal feature extraction and classification method based on modal energy principal component ratio quantification comprises the following steps:
the method comprises the following steps: acquiring a section of normal human body pulse signals through a pulse sensor, wherein the length of the signals is required to be not less than 3 periods; step two: performing CEEMDAN self-adaptive full-integration empirical mode decomposition on the acquired signals, and decomposing original complex signals into modal components with different time scales;
before decomposition, an operator E is defined k (. to solve the k-th modal component IMF of the EMD decomposition k ；
Definition of ω j (t) white noise satisfying the standard normal distribution added in the ith experiment;
definition of ε k The amplitude coefficient of the white noise added for the Kth time;
defining X (t) as an original signal sequence;
step 2.1: for signal X (t) + epsilon 0 ω i (t) performing I times of tests, decomposing by EMD method to obtainA first modal component, the first modal component being calculated by:
step 2.2: calculating a unique residual signal of a first stage, wherein the calculation formula of the unique residual signal is as follows:
r 1 (t)＝X(t)-IMF 1 ；
step 2.3: constructing a signal r 1 (t)+ε 1 E 1 (ω i (t)), then performing EMD, calculating a second modal component, the second modal component being calculated by the formula:
step 2.4: in each subsequent stage, the Kth residual component r is calculated k (t)＝r k-1 (t)-IMF k Constructing a new signal, executing EMD, and obtaining a K +1 mode component, wherein a calculation formula of the K +1 mode component is as follows:
step 2.5: repeating the step 2.4 until the value of the residual component is less than the two extreme values, and stopping decomposition;
finally obtaining the residual component
The original signal is finally decomposed into K modal components, which are expressed as:
step three: calculating a Modal Energy Principal Component Ratio (MEPCR), further processing the decomposed signals, and quantizing the signals for describing the energy principal component distribution of the pulse;
step 3.1: decomposing an original signal into K IMF components through a CEEMDAN algorithm, performing correlation calculation between each component and the original signal, and screening out IMF components close to the original signal, wherein the specific calculation steps are as follows:
correlation coefficient
The calculation formula of (c) is:
wherein Cov (x, IMF) i ) IMF for original sequence and modal components i Covariance of (a) x For the original sequence variance, σ IMF i Is the modal component variance;
the fractions were screened based on the above formula, with the screening rules defined as follows:
{IMF j ：p xIMFi ≧ 0.5}, where j ∈ i, i ═ 1, 2,. k;
it is known that, in signal correlation, a correlation coefficient of 0.5 or more is significant correlation;
step 3.2: the calculation formula of the modal energy principal component ratio MEPCR is as follows:
in the formula:
compared with the prior art, the invention has the beneficial effects that: the invention provides a new pulse parameter to quantify the signal characteristics, so that the pulse parameter is applied to the extraction and classification of pulse signals; the newly proposed pulse parameter MEPCR has good signal characteristic description effect in the aspect of signal energy distribution, and especially plays an important role in extracting and analyzing nonlinear and non-stable pulse signal characteristics; corresponding experiments prove that the new parameter MEPCR has remarkable effect on characterization of the pulse signal features, and the information contained in the pulse signal can be further comprehensively and comprehensively quantified by using a combination mode of the traditional features and the MEPCR in practical application so as to improve the consistency and reliability of the pulse signal diagnosis result.
Drawings
The invention is further described below with reference to the accompanying drawings:
FIG. 1 is a flowchart illustrating a process of extracting pulse signal features according to the present invention;
FIG. 2 is a diagram illustrating the effect of decomposing a pulse signal by an adaptive fully integrated empirical mode according to the present invention;
FIG. 3 is a diagram illustrating the correlation between components and their respective original signals according to an embodiment of the present invention;
FIG. 4 is a graph comparing pre-meal post-meal MEPCR values according to embodiments of the present invention;
FIG. 5 is a scatter plot of MEPCR parameters after a meal for all pulse signals of a data set according to an embodiment of the present invention;
FIG. 6 is an analysis diagram of a classifier model constructed according to an embodiment of the present invention.
Detailed Description
As shown in fig. 1, the present invention relates to the field of signal feature extraction and classification, and proposes a new pulse parameter to quantify a signal feature for accurate classification of a pulse signal, where the pulse parameter is specifically a modal energy principal component ratio MEPCR, and can be applied to pulse signal feature extraction, and the specific steps are as follows:
(1) signal acquisition:
firstly, a section of normal human pulse signals are collected, the length of the signals is required to be 3-5 periods, and in a specific experiment, the pulse signals are collected through an HK-2000C pulse sensor.
(2) Performing CEEMDAN self-adaptive fully-integrated empirical mode decomposition on the acquired signals:
the main function of adopting CEEMDAN decomposition is to decompose original complex signals into modal components with different time scales, and the specific steps of performing CEEMDAN decomposition on the signals are as follows:
first, an operator E is defined k (. the effect of which is to solve the k-th modal component IMF of the EMD decomposition k ；
Let omega i (t) white noise, ε, satisfying the standard normal distribution added in the ith experiment k For the amplitude coefficient of white noise added K times, x (t) is the original signal sequence:
step 2.1: for signal X (t) + epsilon 0 ω i (t) carrying out I times of tests, decomposing by an EMD method to obtain a first modal component, wherein the calculation formula is as follows:
step 2.2: and calculating the unique residual signal of the first stage by the following formula:
r 1 (t)＝X(t)-IMF 1 ；
step 2.3: constructing a signal r 1 (t)+ε 1 E 1 (ω i (t)), then performing EMD to calculate a second modal component, the formula:
step 2.4: at each stage, the Kth residual component r is calculated k (t)＝r k-1 (t)-IMF k Constructing a new signal, and executing EMD, wherein the calculation formula of the K +1 mode component is as follows:
step 2.5: repeating the step 2.4 until the value of the residual component is less than the two extreme values, and stopping decomposition; finally obtaining the residual component
To maximize the original signalThe final decomposition into K modal components is expressed as:
the signal decomposition diagram is shown in fig. 2;
(3) further processing the decomposed signal and quantizing;
when the physiological condition of a human body changes, the energy of certain frequency bands in the human body pulse signal changes, so that the energy distribution is influenced, and therefore the energy characteristics can sensitively detect the tiny change of the pulse signal, and the pulse signal analysis is facilitated.
The invention further provides a new pulse parameter, namely a modal component ratio (MEPCR) for describing the energy principal component distribution of the pulse based on the CEEMDAN theory.
The specific calculation method is as follows:
step 3.1: screening main components: decomposing K IMF components of the original signal through a CEEMDAN algorithm, performing correlation calculation between each component and the original signal, and screening out IMF components close to the original signal.
The specific implementation steps are as follows: calculating a correlation coefficient
The calculation formula is as follows:
wherein Cov (x, IMF) i ) IMF for original sequence and modal components i Covariance of (a) x In order to be the variance of the original sequence,
is the modal component variance. Then, the components are screened according to the following screening rules:
{IMF j ：p xIMFi not less than 0.5}, which isJ ∈ i, i ═ 1, 2,. k;
it is known that a correlation number of 0.5 or more is significant in signal correlation.
Step 3.2: the modal energy principal component ratio MEPCR is:
(4) And (3) experimental verification:
in order to evaluate the effectiveness of the new characteristics, the invention collects and analyzes the pulse signals of the same person after meals under the same measuring condition, and respectively calculates new parameters of the two groups of signals according to the steps, and the specific process is as follows:
firstly, two groups of signals are subjected to CEEMDAN decomposition, the standard deviation of the added noise is 0.2, the noise adding times are 500, the maximum iteration times are 240, and the two groups of signals are decomposed into 8 IMF components and a residual component through CEEMDAN.
Then, MEPCR calculations were performed:
component screening is performed first, and correlations between the two sets of signals and respective modal components are calculated according to the aforementioned method, and the correlations between the two sets of components and respective original signals are shown in fig. 3.
And (4) performing relevance screening on the components according to the screening rule, wherein the IMF components corresponding to the circular labeling part in the figure 3 are screened IMF components.
The table below shows the results of screening two sets of modal components of the signal.
After component screening is completed, MEPCR calculation is carried out according to steps, a comparison graph of the pre-meal and post-meal MEPCR values is shown in FIG. 4, as can be seen from FIG. 4, the MEPCR parameters of the pre-meal and post-meal pulse signals of the same person are obviously different, and the MEPCR of the post-meal pulse signals is obviously higher than the MEPCR parameters of the pre-meal pulse signals.
In order to verify the effectiveness and universality of the feature in the pulse signal classification process, the embodiment constructs a data set containing 200 pulses, the data source is 200 volunteers aged 23-26 years and healthy, and pulse signals of experimenters are respectively collected before and after breakfast, and the specific data set composition information is shown in the following table:
after data acquisition is finished, performing MEPCR parameter calculation on all pulse signals according to steps, and then analyzing results.
FIG. 5 is a scatter plot of the MEPCR parameters of all the pulse signals in the data set, and it can be seen from FIG. 5 that the MEPCR parameters of the human body pre-meal post-meal pulse signals are obviously different and regularly distributed, the MEPCR parameters of the pre-meal pulse signals (star icon) are obviously smaller than the MEPCR values of the post-meal pulse signals (diamond icon), the MEPCR values of the pre-meal pulse signals are distributed between 0.5 and 0.75, and the MEPCR values of the post-meal pulse signals are basically distributed above 0.78. The quantitative effect of the integral characteristics of the MEPCR parameters is obvious.
To further verify the pulse signal validity, the present invention will use the above data set for classifier construction:
firstly, sample marking and label setting are carried out:
wherein the MEPCR q MEPCR value, MEPCR for Pre-meal Signal h MEPCR value as postprandial signal;
subjecting the MEPCR to q Set as tag 1, MEPCR h Set as tag 2;
when the classifier is constructed, a Cubic SVM is selected as a classification model, five-layer cross inspection is set, the model construction and analysis results are shown in fig. 6, a triangular icon is shown in the graph as a label 2, a circular icon is shown in the graph as a label 1, the recognition accuracy of the trained model signal is known to be 95.0%, and the result shows that the new parameter feature MEPCR feature is effective.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (1)
1. A pulse signal feature extraction and classification method based on modal energy principal component ratio quantification is characterized in that: the method comprises the following steps:
the method comprises the following steps: acquiring a section of normal human pulse signals through a pulse sensor, wherein the length of the signals is required to be not less than 3 periods;
step two: performing CEEMDAN self-adaptive full-integration empirical mode decomposition on the acquired signals, and decomposing original complex signals into modal components with different time scales;
before decomposition, an operator E is defined k (. to solve the k-th modal component IMF of the EMD decomposition k ；
Definition of ω i (t) white noise satisfying the standard normal distribution added in the ith experiment;
definition of ε k The amplitude coefficient of the white noise added for the Kth time;
defining X (t) as an original signal sequence;
step 2.1: for signal X (t) + epsilon 0 ω i (t) performing I times of tests, and decomposing by an EMD method to obtain a first modal component, wherein the calculation formula of the first modal component is as follows:
step 2.2: calculating a unique residual signal of a first stage, wherein the calculation formula of the unique residual signal is as follows:
r 1 (t)＝X(t)-IMF 1 ；
step 2.3: constructing a signal r 1 (t)+ε 1 E 1 (ω i (t)), then performing EMD, calculating a second modal component, the second modal component being calculated by the formula:
step 2.4: in each subsequent stage, the Kth residual component r is calculated k (t)＝r k-1 (t)-IMF k Constructing a new signal, executing EMD, and obtaining a K +1 mode component, wherein a calculation formula of the K +1 mode component is as follows:
step 2.5: repeating the step 2.4 until the value of the residual component is less than the two extreme values, and stopping decomposition;
finally obtaining the residual component
The original signal is finally decomposed into K modal components, which are expressed as:
step three: calculating a Modal Energy Principal Component Ratio (MEPCR), further processing the decomposed signals, and quantizing the signals for describing the energy principal component distribution of the pulse;
step 3.1: decomposing an original signal into K IMF components through a CEEMDAN algorithm, performing correlation calculation between each component and the original signal, and screening out IMF components close to the original signal, wherein the specific calculation steps are as follows:
correlationCoefficient of performance
The calculation formula of (c) is:
wherein Cov (x, IMF) i ) IMF for original sequence and modal components i Covariance of (a) x In the form of the variance of the original sequence,
is the modal component variance;
the fractions were screened based on the above formula, with the screening rules defined as follows:
{IMF j ：p xIMFi ≧ 0.5}, where j ∈ i, i ═ 1, 2,. k;
it is known that, in signal correlation, a correlation coefficient of 0.5 or more is significant correlation;
step 3.2: the calculation formula of the modal energy principal component ratio MEPCR is as follows:
in the formula:
the MEPCR value of the pre-meal pulse signal is between 0.5 and 0.75, and the MEPCR value of the post-meal pulse signal is more than 0.78, and the classification is carried out according to the MEPCR value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011035168.8A CN112155523B (en) | 2020-09-27 | 2020-09-27 | Pulse signal feature extraction and classification method based on modal energy principal component ratio quantification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011035168.8A CN112155523B (en) | 2020-09-27 | 2020-09-27 | Pulse signal feature extraction and classification method based on modal energy principal component ratio quantification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112155523A CN112155523A (en) | 2021-01-01 |
| CN112155523B true CN112155523B (en) | 2022-09-16 |
Family
ID=73861361
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011035168.8A Active CN112155523B (en) | 2020-09-27 | 2020-09-27 | Pulse signal feature extraction and classification method based on modal energy principal component ratio quantification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112155523B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1076586A1 (en) * | 1998-05-06 | 2001-02-21 | Exogen, Inc. | Ultrasound bandages |
| EP1482142A2 (en) * | 2001-08-06 | 2004-12-01 | Southwest Research Institute | Method and apparatus for testing catalytic converter durability |
| CN108013548A (en) * | 2018-01-24 | 2018-05-11 | 太原理工大学 | Can pinpoint human body physical sign monitoring Intelligent bracelet |
| EP3669757A1 (en) * | 2018-12-18 | 2020-06-24 | Koninklijke Philips N.V. | System and method for detecting fluid accumulation |
Family Cites Families (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| ZA965340B (en) * | 1995-06-30 | 1997-01-27 | Interdigital Tech Corp | Code division multiple access (cdma) communication system |
| US8388530B2 (en) * | 2000-05-30 | 2013-03-05 | Vladimir Shusterman | Personalized monitoring and healthcare information management using physiological basis functions |
| WO2007083933A1 (en) * | 2006-01-18 | 2007-07-26 | Lg Electronics Inc. | Apparatus and method for encoding and decoding signal |
| US20110245628A1 (en) * | 2010-03-31 | 2011-10-06 | Nellcor Puritan Bennett Llc | Photoplethysmograph Filtering Using Empirical Mode Decomposition |
| US8898037B2 (en) * | 2010-04-28 | 2014-11-25 | Nellcor Puritan Bennett Ireland | Systems and methods for signal monitoring using Lissajous figures |
| DE102010053323B3 (en) * | 2010-12-02 | 2012-05-24 | Xtreme Technologies Gmbh | Method for the spatially resolved measurement of parameters in a cross section of a beam of high-energy, high-intensity radiation |
| US10357163B1 (en) * | 2012-06-01 | 2019-07-23 | Vital Connect, Inc. | Respiratory rate detection using decomposition of ECG |
| CN104414688A (en) * | 2013-08-23 | 2015-03-18 | 北京大学 | Ensemble empirical mode decomposition-based vasovagal syncope precursor detection method |
| CN103496625B (en) * | 2013-10-17 | 2015-05-20 | 太原理工大学 | Multi-rope friction lifter load identification method based on vibration analysis |
| CN104921715A (en) * | 2015-06-09 | 2015-09-23 | 上海华旌科技有限公司 | Multi-parameter vital sign measurement device |
| TWI562758B (en) * | 2015-11-18 | 2016-12-21 | Univ Nat Chiao Tung | Physiological signal measuring system and method thereof |
| IT201700081018A1 (en) * | 2017-07-18 | 2019-01-18 | St Microelectronics Srl | TREATMENT OF ELECTROPHYSIOLOGICAL SIGNALS |
| CN109498041B (en) * | 2019-01-15 | 2021-04-16 | 吉林大学 | Driver road rage state identification method based on electroencephalogram and pulse information |
| CN110090010B (en) * | 2019-06-17 | 2022-04-26 | 北京心数矩阵科技有限公司 | Non-contact blood pressure measuring method and system |
| CN110309817B (en) * | 2019-07-19 | 2020-10-02 | 北京理工大学 | Pulse wave motion artifact removing method for parameter adaptive optimization VMD |
| CN111243739A (en) * | 2020-01-07 | 2020-06-05 | 四川大学 | Anti-interference physiological parameter telemetering method and system |
-
2020
- 2020-09-27 CN CN202011035168.8A patent/CN112155523B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1076586A1 (en) * | 1998-05-06 | 2001-02-21 | Exogen, Inc. | Ultrasound bandages |
| EP1482142A2 (en) * | 2001-08-06 | 2004-12-01 | Southwest Research Institute | Method and apparatus for testing catalytic converter durability |
| CN108013548A (en) * | 2018-01-24 | 2018-05-11 | 太原理工大学 | Can pinpoint human body physical sign monitoring Intelligent bracelet |
| EP3669757A1 (en) * | 2018-12-18 | 2020-06-24 | Koninklijke Philips N.V. | System and method for detecting fluid accumulation |
Non-Patent Citations (2)
| Title |
|---|
| State of Health Monitoring and Remaining Useful Life Prediction of Lithium-Ion Batteries Based on Temporal Convolutional Network;DANHUA ZHOU;《IEEE Access》;20200316;第1-5页 * |
| 光电容积脉搏信号特征提取与分类识别研究;张兴;《中国优秀硕士学位论文全文数据库》;20181231;第1-68页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112155523A (en) | 2021-01-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Nguyen et al. | Adaptive ECG denoising using genetic algorithm-based thresholding and ensemble empirical mode decomposition | |
| Dash et al. | Multi-channel EEG based automatic epileptic seizure detection using iterative filtering decomposition and Hidden Markov Model | |
| CN111310570B (en) | Electroencephalogram signal emotion recognition method and system based on VMD and WPD | |
| Khan et al. | Automated classification of lung sound signals based on empirical mode decomposition | |
| Schuhfried et al. | Classification of 7 monofloral honey varieties by PTR-ToF-MS direct headspace analysis and chemometrics | |
| Güler et al. | A modified mixture of experts network structure for ECG beats classification with diverse features | |
| CN109431497A (en) | A kind of brain-electrical signal processing method and epilepsy detection system | |
| CN107045624B (en) | Electroencephalogram signal preprocessing and classifying method based on maximum weighted cluster | |
| Ellis et al. | A novel local explainability approach for spectral insight into raw eeg-based deep learning classifiers | |
| Zhu et al. | Predicting depression from internet behaviors by time-frequency features | |
| CN112155523B (en) | Pulse signal feature extraction and classification method based on modal energy principal component ratio quantification | |
| CN113069117A (en) | Electroencephalogram emotion recognition method and system based on time convolution neural network | |
| Liu et al. | Evaluating and improving automatic sleep spindle detection by using multi-objective evolutionary algorithms | |
| CN117017297A (en) | Method for establishing prediction and identification model of driver fatigue and application thereof | |
| CN115211870A (en) | Neonate's brain electric signal convulsion discharge detecting system based on multiscale feature fusion network | |
| Lan et al. | Improved wavelet packet noise reduction for microseismic data via fuzzy partition | |
| CN114947850A (en) | Mental load grade objective detection method based on pulse Bouss model characteristics | |
| CN113066544B (en) | FVEP characteristic point detection method based on CAA-Net and LightGBM | |
| CN112200228A (en) | Epileptic seizure state identification method based on two-dimensional convolutional neural network | |
| CN111631711B (en) | Method for analyzing and processing brain electrical data of schizophrenia and auditory hallucination symptoms | |
| CN111631709B (en) | Method for distinguishing auditory hallucination symptoms of schizophrenia from auditory hallucination symptoms of other diseases | |
| CN113796873B (en) | Wearable dynamic electrocardiosignal classification method and system | |
| Ismail et al. | Prediction Model for Spectroscopy Using Python Programming | |
| CN117076868B (en) | Modeling method for persistent data model | |
| Huang et al. | Bandpass empirical mode decomposition using a rolling ball algorithm |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230517 Address after: AW03, AW08, 4th Floor, Building 4, No. 11 Kangshou Street, Tanghuai Industrial Park, Shanxi Transformation and Comprehensive Reform Demonstration Zone, Taiyuan City, Shanxi Province, 030032 (China Shanxi Overseas Students Entrepreneurship Park) Patentee after: Shanxi Hi Tan Ke Technology Co.,Ltd. Address before: 030024 No. 79 West Main Street, Taiyuan, Shanxi, Yingze Patentee before: Taiyuan University of Technology |
|
| TR01 | Transfer of patent right |