CN114391808A - Sleep staging method and system based on nonlinear interdependency - Google Patents
Sleep staging method and system based on nonlinear interdependency Download PDFInfo
- Publication number
- CN114391808A CN114391808A CN202111589919.5A CN202111589919A CN114391808A CN 114391808 A CN114391808 A CN 114391808A CN 202111589919 A CN202111589919 A CN 202111589919A CN 114391808 A CN114391808 A CN 114391808A
- Authority
- CN
- China
- Prior art keywords
- nonlinear
- sleep
- time point
- electroencephalogram
- interdependencies
- 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.)
- Pending
Links
- 230000007958 sleep Effects 0.000 title claims abstract description 88
- 238000000034 method Methods 0.000 title claims abstract description 51
- 230000008667 sleep stage Effects 0.000 claims abstract description 24
- 238000010168 coupling process Methods 0.000 claims description 23
- 238000005859 coupling reaction Methods 0.000 claims description 23
- 230000008878 coupling Effects 0.000 claims description 22
- 210000004556 brain Anatomy 0.000 claims description 14
- 238000004590 computer program Methods 0.000 claims description 14
- 238000003860 storage Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 description 17
- 238000004364 calculation method Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000012360 testing method Methods 0.000 description 6
- 238000001514 detection method Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 238000013459 approach Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 230000000875 corresponding effect Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 239000012634 fragment Substances 0.000 description 3
- 230000002618 waking effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000005314 correlation function Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000037053 non-rapid eye movement Effects 0.000 description 2
- 238000005312 nonlinear dynamic Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000003860 sleep quality Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000000739 chaotic effect Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000002565 electrocardiography Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 230000004461 rapid eye movement Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000029058 respiratory gaseous exchange Effects 0.000 description 1
- 230000033764 rhythmic process Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4806—Sleep evaluation
- A61B5/4812—Detecting sleep stages or cycles
-
- 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
- A61B5/72—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
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Surgery (AREA)
- Public Health (AREA)
- Pathology (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Veterinary Medicine (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Psychiatry (AREA)
- Psychology (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Signal Processing (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention provides a sleep staging method and a system based on nonlinear interdependence, which comprises the steps of obtaining two electroencephalogram signals of a tested person within a certain period of time; extracting various nonlinear interdependencies between the two electroencephalogram signals and cross-correlation coefficients between the two electroencephalogram signals, and combining the correlation coefficients into a sleep characteristic; and based on the sleep characteristics, obtaining the sleep stage of the tested person in the period of time by adopting a classifier. Compared with the traditional linear method, more information can be provided, and the sleep staging accuracy is improved.
Description
Technical Field
The invention belongs to the technical field of electroencephalogram signal processing, and particularly relates to a sleep staging method and system based on nonlinear interdependency.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The detection and the evaluation of the sleep quality have important clinical significance and practical application value, and the sleep state analysis is an important basis for evaluating the sleep quality. In the 30 s of the 20 th century, the german psychiatrist Berger found that the brain electrical (EEG) activity of humans in the sleep and wake phases presented different rhythms. In 1968, the R & K sleep staging criteria were proposed to divide the sleep process into a wake phase (W), a rapid eye movement phase (REM) and a non-rapid eye movement phase (NREM), wherein the non-rapid eye movement phase is divided into phases S1, S2, S3 and S4, and thus the sleep staging problem is a six-class problem. The traditional sleep staging is judged by means of naked eyes of experts, and the process is time-consuming and has subjective judgment probability, so that the traditional sleep analysis method is low in accuracy rate of the sleep staging.
In recent years, studies have been increasingly made on extracting features based on physiological signals such as single-channel EEG, multi-channel EEG, electrocardiography, electrooculogram, myoelectricity, and respiration, and performing sleep stages using classifiers. The sleep stage effect based on the EEG is the best, the electrocardio is the second time, and the myoelectricity and electrooculogram effects are poor. However, existing EEG-based sleep staging methods rely on a priori assumptions and are not highly accurate and slow to sleep stage.
Disclosure of Invention
In order to solve the technical problems in the background art, the invention provides a sleep staging method and system based on nonlinear interdependence, which extracts nonlinear measurement as a characteristic and is applied to sleep detection.
In order to achieve the purpose, the invention adopts the following technical scheme:
a first aspect of the present invention provides a sleep staging method based on non-linear interdependencies, comprising:
acquiring two electroencephalogram signals of a tested person within a certain period of time;
extracting various nonlinear interdependencies between the two electroencephalogram signals and cross-correlation coefficients between the two electroencephalogram signals, and combining the correlation coefficients into a sleep characteristic;
and based on the sleep characteristics, obtaining the sleep stage of the tested person in the period of time by adopting a classifier.
Further, the specific steps of extracting multiple nonlinear interdependencies between two electroencephalogram signals and the cross-correlation coefficient between the two electroencephalogram signals are as follows:
reconstructing each electroencephalogram signal to obtain reconstructed electroencephalogram signals;
calculating the independent neighborhood distance and the coupling neighborhood distance of each time point in the reconstructed electroencephalogram signal;
based on the independent neighborhood distance and the coupling neighborhood distance of each time point in the two electroencephalogram signals, various nonlinear interdependencies between the two electroencephalogram signals and cross-correlation coefficients between the two electroencephalogram signals are calculated.
Further, the independent neighborhood distance is a mean square euclidean distance from a certain time point in one electroencephalogram signal to a k neighborhood time point of the time point in the same electroencephalogram signal.
Further, the coupling neighborhood distance is a mean square euclidean distance from a certain time point in one electroencephalogram signal to a k neighborhood time point of the time point in another electroencephalogram signal.
Further, the plurality of nonlinear interdependencies include a first nonlinear interdependency, and the calculation method of the first nonlinear interdependency is as follows: and calculating the ratio of the independent neighborhood distance to the coupling neighborhood distance of the electroencephalogram signal at each time point, calculating the mean value of the ratio at all the time points, and taking the mean value as a first nonlinear interdependence degree.
Further, the plurality of nonlinear interdependencies include a second nonlinear interdependency, and the calculation method of the second nonlinear interdependency is as follows: and calculating the ratio of the independent neighborhood distance to the coupling neighborhood distance of the electroencephalogram signal at each time point, calculating the logarithm value of the ratio at each time point, and taking the mean value of the logarithm values at all the time points as a second nonlinear interdependence.
Further, the plurality of nonlinear interdependencies include a third nonlinear interdependency, and the calculation method of the third nonlinear interdependency is: calculating to obtain a difference value between the independent neighborhood distance and the coupling neighborhood distance of the electroencephalogram signal at each time point, calculating a ratio of the difference value to the independent neighborhood distance at the same time point, and taking the mean value of the ratios at all time points as a third nonlinear interdependence.
A second aspect of the present invention provides a sleep staging system based on non-linear interdependencies, comprising:
a signal acquisition module configured to: acquiring two electroencephalogram signals of a tested person within a certain period of time;
a feature extraction module configured to: extracting various nonlinear interdependencies between the two electroencephalogram signals and cross-correlation coefficients between the two electroencephalogram signals, and combining the correlation coefficients into a sleep characteristic;
a classification module configured to: and based on the sleep characteristics, obtaining the sleep stage of the tested person in the period of time by adopting a classifier.
A third aspect of the present invention provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps in a non-linear interdependence based sleep staging method as described above.
A fourth aspect of the present invention provides a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of a non-linear interdependence based sleep staging method as described above when executing the program.
Compared with the prior art, the invention has the beneficial effects that:
the invention provides a sleep stage method based on nonlinear interdependency, which applies the nonlinear interdependency to sleep stage, does not need to extract other characteristics and variables in electroencephalogram signals, only needs time sequence of electroencephalogram signals, and is embodied more directly; the nonlinear interdependence measurement is the coupling strength between dynamic systems, and when the sleep period changes, the electroencephalogram signal changes violently, so that strong nonlinear characteristics are reflected, and at the moment, the nonlinear dependency shows strong fluctuation; the nonlinear interdependence uses time sequence data of the electroencephalogram signals, is convenient to acquire and monitor, and has better timeliness.
The invention provides a sleep staging method based on nonlinear interdependency, which adopts a fuzzy logic classifier, wherein the classifier can identify a prototype from observation data and construct a 0-order Anya type fuzzy rule; the element parameters obtained by the method are directly derived from data and do not depend on prior hypothesis; moreover, the performance and the computational efficiency can be balanced according to the adjustment of the computational complexity, and meanwhile, the fuzzy logic classifier also supports different types of distance metrics, so that the fuzzy logic classifier can be efficiently adjusted according to specific problems.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
Fig. 1 is a flowchart of a sleep staging method based on nonlinear interdependencies according to a first embodiment of the present invention.
Detailed Description
The invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
Example one
The embodiment provides a sleep staging method based on nonlinear interdependence, as shown in fig. 1, specifically including the following steps:
step 1, acquiring two electroencephalogram signals of a tested person within a certain period of time: first electroencephalogram signal X '═ { X'nAnd a second electroencephalogram signal Y '{ Y'nAnd (c) the step of (c) in which,n＝1，…，N。
specifically, the method comprises the following steps: (101) acquiring two electroencephalogram signals of a tested person, wherein different electroencephalogram signals are acquired by adopting different electroencephalogram lead methods; supposing that two electroencephalogram lead methods x and y exist, a first electroencephalogram signal obtained by adopting the first electroencephalogram lead method x and a second electroencephalogram signal obtained by adopting the second electroencephalogram lead method y; (102) segmenting each electroencephalogram signal to obtain an electroencephalogram signal in each period of time; in particular, according to R&According to the K rule, each electroencephalogram signal is divided into 30 second segments, and a first electroencephalogram signal X ' ═ { X ' in each segment is obtained 'nAnd a second electroencephalogram signal Y '{ Y'nWhere N is 1, …, N.
Or two electroencephalogram signals of the tested person within a period of time (30 seconds) are obtained by directly adopting two electroencephalogram lead methods.
Thus, each segment of brain electrical signal X 'or Y' corresponds to a sleep stage (W, REM, S1, S2, S3, or S4).
And 2, extracting various nonlinear interdependencies between the two electroencephalogram signals and cross correlation coefficients between the two electroencephalogram signals, and combining the correlation coefficients into a sleep characteristic.
(201) And reconstructing each electroencephalogram signal to obtain the reconstructed electroencephalogram signal.
Specifically, from two electroencephalogram leads X and y, a time sequence X '═ { X'nY'nReconstructing the electroencephalogram signal, wherein the delay vector of each time point in the reconstructed electroencephalogram signal is xn＝(x′n,…,x′n-(m-1)τ) And yn＝(y′n,…,y′n-(m-1)τ) N is 1, … N, where N is the time point and x isnAnd representing a delay vector of an nth time point in the first brain electrical signal, m is an embedding dimension, and tau represents a time lag.
Therefore, the reconstructed first brain electrical signal X ═ { X ═ XnAnd the reconstructed second electroencephalogram signal Y is equal to { Y ═ Y }n}。
(202) And calculating the independent neighborhood distance of the delay vector of each time point in the reconstructed electroencephalogram signal. The independent neighborhood distance is the mean square Euclidean distance from a certain time point in one electroencephalogram signal to the k neighborhood time point of the time point in the same electroencephalogram signal.
Specifically, based on the delay vector X of each time point in the reconstructed first electroencephalogram signal XnCalculating a certain time point x in an electroencephalogram signalnTo the time point x in the same EEG signalnK neighborhood of (a), i.e., each time point x in the first brain electrical signalnThe independent neighborhood distance of (c) is:
similarly, the delay vector Y for each time point in the reconstructed second EEG signal YnThe delay vector y for each time point can be calculatednThe mean square Euclidean distance from k neighborhood of the first electroencephalogram to obtain a delay vector y of each time point in the second electroencephalogram signalnHas an independent neighborhood distance of
(203) And calculating the coupling neighborhood distance of the delay vector of each time point in the reconstructed electroencephalogram signal. The coupling neighborhood distance is the mean square Euclidean distance from a certain time point in one electroencephalogram signal to the k neighborhood time point of the time point in the other electroencephalogram signal.
Delay vector X for each time point in the reconstructed first brain electrical signal XnCalculating the delay vector x of each time point under the condition of the reconstructed second electroencephalogram signal YnMean squared Euclidean distance to k neighborhood of the time point in another brain electrical signal, the distance being defined by replacing the nearest neighbor with an equal time neighbor of the y nearest neighbor, i.e., a delay vector x for each time point in the first brain electrical signalnCoupling neighborhood distance of (c):
wherein r isn,jAnd sn,jJ is 1, …, k each represents xnAnd ynK is the time index of the nearest neighbor.
Similarly, the delay vector Y for each time point in the reconstructed second EEG signal YnCalculating the delay vector y of each time point under the condition of the reconstructed second brain electrical signal XnThe mean square Euclidean distance from k neighborhood of the first electroencephalogram to obtain a delay vector y of each time point in the second electroencephalogram signalnCoupled neighborhood distance of
(204) Based on the independent neighborhood distance and the coupling neighborhood distance of each time point delay vector in the two electroencephalogram signals, various nonlinear interdependencies between the two electroencephalogram signals and cross-correlation coefficients between the two electroencephalogram signals are calculated. Wherein the sleep characteristics (first sleep characteristic, second sleep characteristic, third sleep characteristic, fourth sleep characteristic, fifth sleep characteristic, sixth sleep characteristic, and seventh sleep characteristic). The multiple nonlinear interdependencies comprise a first nonlinear interdependency S, a second nonlinear interdependency H and a third nonlinear interdependency N, wherein the first nonlinear interdependency S corresponds to a first sleep characteristic and a second sleep characteristic, the second nonlinear interdependency H corresponds to a third sleep characteristic and a fourth sleep characteristic, the third nonlinear interdependency N corresponds to a fifth sleep characteristic and a sixth sleep characteristic, and a cross-correlation coefficient between the two electroencephalogram signals corresponds to a seventh sleep characteristic.
(a) The first nonlinear interdependence S is calculated by the following method: and calculating the ratio of the independent neighborhood distance to the coupling neighborhood distance of the electroencephalogram signal at each time point, calculating the mean value of the ratio at all the time points, and taking the mean value as a first nonlinear interdependence degree.
If the reconstructed first brain electrical signal X is the lattice { XnHas an average square radius
Then if these systems are strongly correlated, there is
If they are independent, then
Thus, an interdependent metric (first sleep characteristic) S can be defined(k)(X | Y) is
Due to the fact that
Is provided with
0<S(k)(X∣Y)≤1
If S is(k)The value of (X | Y) approaches 0, then the relationship between XY is independent, and when S is close to 0(k)When the value of (X | Y) approaches 1, it means that XY will reach the maximum value at the same time.
Similarly, a second sleep characteristic S may be obtained(k)(Y∣X)。
(b) And a second nonlinear interdependence H, wherein the calculation method of the second nonlinear interdependence is as follows: and calculating the ratio of the independent neighborhood distance to the coupling neighborhood distance of the electroencephalogram signal at each time point, calculating the logarithm value of the ratio at each time point, and taking the mean value of the logarithm values at all the time points as a second nonlinear interdependence.
Another metric (third sleep characteristic) H that is not linearly interdependent(k)(X | Y) is defined as
If X and Y are completely independent, then H(k)The value of (X | Y) is 0 and if proximity in Y also means proximity to a peer time partner in X, it will be positive.
Similarly, a fourth sleep characteristic H may be obtained(k)(Y∣X)。
(c) And a third nonlinear interdependence N, wherein the calculation method of the third nonlinear interdependence comprises the following steps: calculating to obtain a difference value between the independent neighborhood distance and the coupling neighborhood distance of the electroencephalogram signal at each time point, calculating a ratio of the difference value to the independent neighborhood distance at the same time point, and taking the mean value of the ratios at all time points as a third nonlinear interdependence.
In previous coupled chaotic system studies H was more robust to noise, but the disadvantage was that it was not normalized, and therefore a new measurement (fifth sleep characteristic) N (X | Y) was proposed, which was normalized and more robust than S.
Similarly, a sixth sleep characteristic N may be obtained(k)(Y∣X)。
In general, S (X | Y), H (X | Y), N (X | Y) are not equal to S (Y | X), H (Y | X), N (Y | X). The asymmetry of S, H, N is a major advantage of other non-linear metrics mutual information and phase synchronization.
(d) Using S (X | Y), H (X | Y), N (X | Y), S (Y | X), H (Y | X) and N (Y | X) as characteristics to make extraction, and in addition, adding cross-correlation function (seventh sleep characteristic) cxyThe cross-correlation function is the most common measure for measuring the difference between two brain electrical signals, here
And σxRepresents { xnThe mean and the variance of the mean and the variance,
and σyRepresents { ynMean and variance of, τ denotes the time lag, and has cxy＝cyx。
In summary, the first sleep characteristic, the second sleep characteristic, the third sleep characteristic, the fourth sleep characteristic, the fifth sleep characteristic, the sixth sleep characteristic, and the seventh sleep characteristic together constitute the sleep characteristic of the subject within a certain period of time.
And 3, obtaining the sleep stage of the tested person in the period of time by adopting a classifier based on the sleep characteristics.
As an embodiment, the classifier employs a fuzzy logic classifier.
When the set of sleep characteristics of the tested person is reached in a certain period of time, the emission intensity is calculated by using a, and the label at the maximum value is selected as a final result:
tag ═ arg max (λ)M(a))
Where M is { W, REM, S1, S2, S3, S4}, a is the test sample, P is the final prototype formed for each class, P is divided into six classes, and each class of sleep stage produces a set of prototypes. The sleep stage is classified into 6 classes, for each class, six prototypes are respectively generated, namely { W, REM, S1, S2, S3 and S4}, when a test sample arrives, the attack intensity between the test sample and the six prototypes is respectively calculated, a certain class of prototypes is not only one, so the intensity in the class is different, the maximum emission intensity in the class is taken as a representative of the test sample and the prototype, so six representatives exist, then the maximum value is taken from the six representatives as the final label, and therefore the maximum value is taken twice. The maximum value is taken in a certain class, and then the maximum values of the six classes are compared.
The fuzzy logic classifier uses non-parametric EDA quantities to objectively reveal the integration properties and mutual distribution of the data, specifically, the following three EDA quantities are employed:
assume that there is a training sample set a ═ a1,a2,…,akAnd the unique data sample set corresponding to the training sample set a is U ═ U }1,u2,…,uUk}：
Cumulative approach, data sample ai(i.e., the sleep characteristics of a subject over a period of time) is represented as:
wherein d (a)i,aj) Denotes ai,ajThe distance between them.
② data sample aiSingle mode density of (a):
③ unique data sample uiMulti-modal density of (a):
wherein u isiFor the unique data sample, fi is the frequency of occurrence of the unique data sample ui. For example, for the training sample set a ═ {1,1,2,3,4,4}, then its unique data sample set U ═ 1,2,3,4}, and the corresponding frequencies fi are 1/3,1/6,1/6,1/3, respectively.
The recursively calculated form of the non-parametric EDA quantity plays an important role in the processing of data. The classifier can be classified by adopting three classification distances, which are respectively: mahalanobis distance, euclidean distance, cosine similarity. Under the three distances, the key element parameters can be stored in the memory to realize quick calculation, and the high efficiency of the method is ensured.
As an embodiment, the fuzzy logic classifier is trained by two training methods: off-line training and on-line training:
(1) off-line training
First, a training sample set is calculated
Each corresponding unique data sample wiMulti modal density of
And will be
The ranking into a list w is performed. Then selects local maximum in the list { w }, then stores the local maximum in { p }0. Take training data of class S1 as an example.
THEN(wi∈{p}0)
Thereafter, { p }0To adsorb the most recent other data samples to form a data cloud. Wp is { p }0The prototype of the data cloud formed after the p-th element in (b) adsorbs the latest data sample, i.e. the prototype of the p-th data cloud:
and then finding the center of the p-th data cloud, and forming a neighborhood with the centers of other data clouds according to the average radius, wherein the data cloud is the neighborhood of the p-th data cloud if the distance between the center of a certain data cloud and the center of the p-th data cloud is smaller than the average radius. Mean radius
Determined by the particle size (L). Generally, the larger the L, the finer the classification.
If the multimodal density of the pth data cloud center is greater than the multimodal density of all data cloud centers in the neighborhood of the pth data cloud, the prototype of the pth data cloud is determined to be a prototype, and the prototype of the pth data cloud is classified as { p }S1。
(2) On-line training
When a new training data arrives, the average radius is first updated
The new training data monomodal densities are then calculated. If the single-mode density is larger than the maximum single-mode density or smaller than the minimum single-mode density of the existing prototype, the prototype becomes a new prototype;
if not, then calculate its distance to the nearest prototype and average radius
A comparison is made. If it is larger than the average distance, it will also become a new prototype.
When the prototype becomes a new prototype, the parameters of the classifier are updated. If it cannot be called a new prototype, it will be included in the most recent prototype while updating the parameters.
Repeating the prototype forming steps for six times to obtain six types of prototypes, which are respectively expressed as { p }REM，{p}S1，{p}s2And the like.
And 4, after obtaining the sleep stages of the tested person in a plurality of continuous time periods, performing post-processing: for six stages of sleep staging, judging each stage in classification, for example, for the waking period, dividing the result into the waking period and the non-waking period, carrying out one-time judgment and obtaining the result. The determination is performed for each of the six stages of the object, and the average value is taken as the final result. Specifically, the sleep stages in h time slots are post-processed, and if more than threshold time slots are in a certain sleep stage in the 1 st time slot to the h1 th time slot and the 1 st time slot and the h1 th time slot are also in the sleep stage, the time slots which are not in the sleep stage in the 1 st time slot to the h1 th time slot are modified to be in the sleep stage. For example, the sleep stages in 400 time slices are post-processed, wherein the 1 st to 100 th slices are S1, and the others are S2, S3 or other stages such as REM. When the judgment is carried out, after a 400-segment result is obtained, the result judgment is carried out on the S1, and the two classification is changed into the S1 type and the non-S1 type. The 1 st to 100 th paragraphs should be judged as S1 type, and the 101 st to 400 th paragraphs should be judged as non-S1 type. If the 20 th fragment is determined to be of class S1, then it is correct; the 200 th fragment is judged as S1 being wrong; the 300 th fragment is judged not to be stage S1, and it is correct. Interpretation is performed on these 400 segments, and the result of S1 is obtained.
The method of the invention is verified on a sleep database provided by the Hospital of san Wensen, Dublin. The database contains a plurality of physiological information records such as electroencephalogram signals, electrocardiosignals and the like. Wherein, C3-A2 and C4-A1 are two electroencephalogram signals, so the two leads are taken as samples, and the nonlinear interdependence relation between the two leads is extracted. In this experiment, the time lag τ is selected to be 2, the embedding dimension m is selected to be 10, the neighborhood K is selected to be 10, and the taylor correction T is selected to be 50. Through selection and experimental test of three classification distances, granularity and training set on-line and off-line training proportion, the method is obtained: and when the cosine similarity and the granularity of the classification distance are 12, and all the training sets are used for off-line training, the accuracy is highest.
Through the experimental results in table 1, it can be seen that the average value of the accuracy rate in the test of five testees reaches 81%, and compared with other methods, the result is better.
TABLE 1 results of the experiment
Based on nonlinear dynamics, different leads for extracting electroencephalogram signals when a subject sleeps can be regarded as a plurality of nonlinear power systems; regarding various connections and differences among the leads of the multichannel electroencephalogram signals as coupling relations among several dynamic systems; because the electroencephalogram signals change continuously between different sleep stages, the coupling relation between different dynamic systems also changes continuously; therefore, the continuous change during the sleep period is comprehensively reflected as the change of the coupling relation between the two brain electrical leads, and the invention adopts a nonlinear dynamics method to research the sleep period. The invention carries out sleep staging by extracting features from the multi-channel electroencephalogram signals, and because many features of the electroencephalogram signals cannot be generated by a linear model, the invention extracts nonlinear measurement as the features, is applied to sleep detection and provides more information than the traditional linear method.
The fuzzy logic classifier is adopted, the classifier does not depend on any prior hypothesis, only depends on the internal relation among data to carry out identification and classification, only stores key element parameters in a memory, has extremely high calculation speed, and solves the problem of low calculation speed of other methods; the invention also has three classification distances for the flexible selection of practical conditions: mahalanobis distance, euclidean distance, cosine similarity; the defect of low accuracy of the prior sleep staging is overcome, and the speed is high; and the computational complexity can be flexibly adjusted: not only can obtain higher detection rate, but also can avoid overfitting caused by overfeeding of the training set; by means of flexible adjustment of the parameters, excellent results are obtained in different situations.
Example two
The embodiment provides a sleep staging system based on nonlinear interdependence, which specifically comprises the following modules:
a signal acquisition module configured to: acquiring two electroencephalogram signals of a tested person within a certain period of time;
a feature extraction module configured to: extracting various nonlinear interdependencies between the two electroencephalogram signals and cross-correlation coefficients between the two electroencephalogram signals, and combining the correlation coefficients into a sleep characteristic;
a classification module configured to: and based on the sleep characteristics, obtaining the sleep stage of the tested person in the period of time by adopting a classifier.
It should be noted that, each module in the present embodiment corresponds to each step in the first embodiment one to one, and the specific implementation process is the same, which is not described herein again.
EXAMPLE III
The present embodiment provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps in a non-linear interdependence based sleep staging method as described in the first embodiment above.
Example four
This embodiment provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the program, the processor implements the steps in the sleep staging method based on nonlinear interdependencies as described in the first embodiment.
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 a hardware embodiment, a 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, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is 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 flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams 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.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A sleep staging method based on nonlinear interdependencies, comprising:
acquiring two electroencephalogram signals of a tested person within a certain period of time;
extracting various nonlinear interdependencies between the two electroencephalogram signals and cross-correlation coefficients between the two electroencephalogram signals, and combining the correlation coefficients into a sleep characteristic;
and based on the sleep characteristics, obtaining the sleep stage of the tested person in the period of time by adopting a classifier.
2. The sleep stage classification method based on nonlinear interdependence as claimed in claim 1, wherein the specific steps of extracting various nonlinear interdependences between two electroencephalogram signals and cross-correlation coefficients between two electroencephalogram signals are as follows:
reconstructing each electroencephalogram signal to obtain reconstructed electroencephalogram signals;
calculating the independent neighborhood distance and the coupling neighborhood distance of each time point in the reconstructed electroencephalogram signal;
based on the independent neighborhood distance and the coupling neighborhood distance of each time point in the two electroencephalogram signals, various nonlinear interdependencies between the two electroencephalogram signals and cross-correlation coefficients between the two electroencephalogram signals are calculated.
3. The method of claim 2, wherein the independent neighborhood distance is the mean squared euclidean distance from a time point in one electroencephalogram signal to a time point in k neighborhood of the time point in the same electroencephalogram signal.
4. The method of claim 2, wherein the coupling neighborhood distance is the mean squared euclidean distance from a time point in one brain electrical signal to a time point k neighborhood of the time point in another brain electrical signal.
5. The sleep staging method based on nonlinear interdependencies as recited in claim 2, wherein the plurality of nonlinear interdependencies include a first nonlinear interdependency calculated by: and calculating the ratio of the independent neighborhood distance to the coupling neighborhood distance of the electroencephalogram signal at each time point, calculating the mean value of the ratio at all the time points, and taking the mean value as a first nonlinear interdependence degree.
6. The sleep staging method based on nonlinear interdependencies as recited in claim 2, wherein the plurality of nonlinear interdependencies includes a second nonlinear interdependency calculated by: and calculating the ratio of the independent neighborhood distance to the coupling neighborhood distance of the electroencephalogram signal at each time point, calculating the logarithm value of the ratio at each time point, and taking the mean value of the logarithm values at all the time points as a second nonlinear interdependence.
7. The sleep staging method based on nonlinear interdependencies as recited in claim 2, wherein the plurality of nonlinear interdependencies includes a third nonlinear interdependency calculated by: calculating to obtain a difference value between the independent neighborhood distance and the coupling neighborhood distance of the electroencephalogram signal at each time point, calculating a ratio of the difference value to the independent neighborhood distance at the same time point, and taking the mean value of the ratios at all time points as a third nonlinear interdependence.
8. A sleep staging system based on non-linear interdependencies, comprising:
a signal acquisition module configured to: acquiring two electroencephalogram signals of a tested person within a certain period of time;
a feature extraction module configured to: extracting various nonlinear interdependencies between the two electroencephalogram signals and cross-correlation coefficients between the two electroencephalogram signals, and combining the correlation coefficients into a sleep characteristic;
a classification module configured to: and based on the sleep characteristics, obtaining the sleep stage of the tested person in the period of time by adopting a classifier.
9. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of a non-linear interdependence based sleep staging method as claimed in any one of the claims 1-7.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the steps in a non-linear interdependence based sleep staging method as claimed in any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111589919.5A CN114391808A (en) | 2021-12-23 | 2021-12-23 | Sleep staging method and system based on nonlinear interdependency |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111589919.5A CN114391808A (en) | 2021-12-23 | 2021-12-23 | Sleep staging method and system based on nonlinear interdependency |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114391808A true CN114391808A (en) | 2022-04-26 |
Family
ID=81227642
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111589919.5A Pending CN114391808A (en) | 2021-12-23 | 2021-12-23 | Sleep staging method and system based on nonlinear interdependency |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114391808A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110092831A1 (en) * | 2008-04-14 | 2011-04-21 | Itamar Medical Ltd. | Non invasive method and apparatus for determining light-sleep and deep-sleep stages |
| CN108968915A (en) * | 2018-06-12 | 2018-12-11 | 山东大学 | Sleep state classification method and system based on entropy feature and support vector machines |
| US20200338304A1 (en) * | 2018-01-16 | 2020-10-29 | Walter Viveiros | System and method for sleep environment management |
| US20200346016A1 (en) * | 2019-05-02 | 2020-11-05 | Enhale Medical, Inc. | Systems and methods to improve sleep disordered breathing using closed-loop feedback |
-
2021
- 2021-12-23 CN CN202111589919.5A patent/CN114391808A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110092831A1 (en) * | 2008-04-14 | 2011-04-21 | Itamar Medical Ltd. | Non invasive method and apparatus for determining light-sleep and deep-sleep stages |
| US20200338304A1 (en) * | 2018-01-16 | 2020-10-29 | Walter Viveiros | System and method for sleep environment management |
| CN108968915A (en) * | 2018-06-12 | 2018-12-11 | 山东大学 | Sleep state classification method and system based on entropy feature and support vector machines |
| US20200346016A1 (en) * | 2019-05-02 | 2020-11-05 | Enhale Medical, Inc. | Systems and methods to improve sleep disordered breathing using closed-loop feedback |
Non-Patent Citations (3)
| Title |
|---|
| HAN G. JO 等: "Genetic fuzzy classifier for sleep stage identification", 《COMPUTERS IN BIOLOGY AND MEDICINE》, vol. 40, no. 7, 31 July 2010 (2010-07-31), XP027141017 * |
| SERAP AYDIN 等: "Mutual Information Analysis of Sleep EEG in Detecting Psycho-Physiological Insomnia", 《EDUCATION & TRAINING》, vol. 39, no. 43, 3 March 2015 (2015-03-03) * |
| 闫彦 等: "轻度认知障碍老年人脑电的非线性相互依赖脑网络分析", 《中国生物医学工程学报》, vol. 40, no. 6, 20 December 2021 (2021-12-20) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Detection analysis of epileptic EEG using a novel random forest model combined with grid search optimization | |
| CN110897639A (en) | Electroencephalogram sleep staging method based on deep convolutional neural network | |
| Diykh et al. | EEG sleep stages identification based on weighted undirected complex networks | |
| Shahnaz et al. | Emotion recognition based on wavelet analysis of Empirical Mode Decomposed EEG signals responsive to music videos | |
| Kaleem et al. | Patient-specific seizure detection in long-term EEG using signal-derived empirical mode decomposition (EMD)-based dictionary approach | |
| Kini et al. | Large margin mixture of AR models for time series classification | |
| Zhu et al. | An efficient visibility graph similarity algorithm and its application on sleep stages classification | |
| CN111631682B (en) | Physiological characteristic integration method and device based on trending analysis and computer equipment | |
| EP3226751A1 (en) | Sleep profiling system with feature generation and auto-mapping | |
| Roy et al. | Photoplethysmogram signal quality evaluation by unsupervised learning approach | |
| Al-Salman et al. | An efficient approach for EEG sleep spindles detection based on fractal dimension coupled with time frequency image | |
| De Falco et al. | Deep neural network hyper-parameter setting for classification of obstructive sleep apnea episodes | |
| CN112263218A (en) | Sleep staging method and device | |
| Putra et al. | EEG-based emotion classification using wavelet decomposition and K-nearest neighbor | |
| Tigga et al. | Efficacy of novel attention-based gated recurrent units transformer for depression detection using electroencephalogram signals | |
| Ohannesian et al. | Epileptic seizures detection from EEG recordings based on a hybrid system of Gaussian mixture model and random forest classifier | |
| CN114391808A (en) | Sleep staging method and system based on nonlinear interdependency | |
| Shi et al. | Automatic sleep stage classification based on LSTM | |
| Taha et al. | EEG Emotion Recognition Via Ensemble Learning Representations | |
| CN113907770A (en) | Method and system for detecting and identifying spike-slow complex wave based on feature fusion | |
| Li et al. | Detection of Obstructive Sleep Apnea Based on ECG System Dynamics, Hybrid Signal Processing and Neural Networks | |
| Ergin et al. | Emotion detection using EEG signals based on Multivariate Synchrosqueezing Transform and Deep Learning | |
| Dlask et al. | Fractional Brownian Bridge Model for Alzheimer Disease Detection from EEG Signal | |
| Liu et al. | Research on feature fusion for emotion recognition based on discriminative canonical correlation analysis | |
| CN114680904B (en) | Electroencephalogram signal quality assessment method based on characteristic wave detection and stage 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 |