CN114569139B - Method for constructing complex brain muscle interaction closed loop function network framework - Google Patents

Method for constructing complex brain muscle interaction closed loop function network framework Download PDF

Info

Publication number
CN114569139B
CN114569139B CN202210176343.8A CN202210176343A CN114569139B CN 114569139 B CN114569139 B CN 114569139B CN 202210176343 A CN202210176343 A CN 202210176343A CN 114569139 B CN114569139 B CN 114569139B
Authority
CN
China
Prior art keywords
muscle
brain
coherent
channels
closed
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
Application number
CN202210176343.8A
Other languages
Chinese (zh)
Other versions
CN114569139A (en
Inventor
刘金标
魏依娜
冯琳清
唐弢
王丽婕
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang Lab
Original Assignee
Zhejiang Lab
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Zhejiang Lab filed Critical Zhejiang Lab
Priority to CN202210176343.8A priority Critical patent/CN114569139B/en
Publication of CN114569139A publication Critical patent/CN114569139A/en
Application granted granted Critical
Publication of CN114569139B publication Critical patent/CN114569139B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/369—Electroencephalography [EEG]
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/369—Electroencephalography [EEG]
    • A61B5/372—Analysis of electroencephalograms
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/389—Electromyography [EMG]
    • 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
    • A61B5/389—Electromyography [EMG]
    • A61B5/397—Analysis of electromyograms
    • 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
    • A61B5/7253—Details of waveform analysis characterised by using transforms
    • A61B5/7257—Details of waveform analysis characterised by using transforms using Fourier transforms
    • 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

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Biophysics (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Pathology (AREA)
  • Psychiatry (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physiology (AREA)
  • Signal Processing (AREA)
  • Psychology (AREA)
  • Fuzzy Systems (AREA)
  • Evolutionary Computation (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

The invention provides a method for constructing a complex brain muscle interaction closed-loop function network framework, and relates to the field of multi-mode information fusion. According to the invention, through brain function network expansion, a cortex-muscle-cortex closed-loop topological network is constructed to explain the change of a cooperative working mode in a human body motion control system, and meanwhile, the directional characteristic for distinguishing causal relationships is given by using an expanded bias orientation coherent method, so that the functional connectivity and effective connectivity of the closed-loop network are established, the dynamic evolution process of the cortex network related to muscle activity is objectively described in a graphic visualization manner, and the application value of electrophysiological information in the evaluation of the upper limb motion dysfunction of a patient is greatly improved.

Description

Method for constructing complex brain muscle interaction closed-loop function network framework
Technical Field
The invention relates to the field of multi-mode information fusion, in particular to a method for constructing a closed-loop functional network framework of cerebral cortex-muscle information interaction.
Background
The connectivity analysis of the cerebral cortex network has become an important tool for researching the brain function in the process of recovering the motor function of the patient with limb disorder. A great deal of research has been directed to exploiting functional or effective connectivity of neuroimaging and electrophysiological data to explore the interactions between different areas of the macroscopic brain and its effects on behavior, and to demonstrate that changes in brain function are associated with loss and recovery of motor function. This systematic view of the brain network allows an understanding of the pathophysiology of dyskinesias with a topological nature, which may therefore influence therapeutic strategies for intervening pathological brain networks.
Despite the great progress made in studies on brain network connectivity with impaired motor function, little is known about the dynamic changes of the cortical network associated with muscle activity, and knowledge of these changes can enhance the overall understanding of the changes in the cooperative working patterns in the motor control system of the patient.
The motion is generated as a result of the coordinated operation of the entire sensorimotor system and is coupled in a complex feedback loop. While focusing on the cortical network, the role of effector muscles in motor control cannot be easily overlooked. The functional coupling of brain muscles can reflect the interaction between cerebral cortex and muscle activity, the coupling between muscles has been used as a supplementary means for detecting the change of corticospinal tracts, both are the characteristics of coupling estimation methods, and the methods mainly utilize various coherent models to quantify recorded brain electrical signals and muscle electrical signals so as to partially describe the working mechanism of a motion control system.
However, no research has been done in the current stage to integrate the information interaction between cortex and muscle into the cortical network, and to explore the potential mechanism change related to the motor function of the limb disorder patient in a closed loop system manner.
Disclosure of Invention
In view of the above-mentioned defects of the existing cortical network technology, the present invention provides a new neural network framework construction method to study the dynamic operation mode of the cortical-muscle-cortical network topology of the patient.
The technical scheme adopted by the invention is as follows:
a method for constructing a complex brain-muscle interaction closed-loop functional network framework comprises the following steps:
step 1: combining the acquired and recorded electroencephalogram (EEG) data and Electromyogram (EMG) data into an N + M dimensional time sequence; wherein, N is the channel number of the electroencephalogram data, and M is the channel number of the electromyogram data.
Step 2: describing the time series as a multiple autoregressive process, as follows:
Figure BDA0003520415100000021
wherein Xr(t) is a time series, AkFor sparse matrices, p represents the model order, k represents the lag variable, εr(t) representing white noise, and expanding the white noise into a transient effect and a hysteresis effect by setting a value of 0 or non-0 according to a hysteresis variable k to obtain an initial model;
and 3, step 3: estimating the optimal model order of the initial model based on Akaike information standard, and correcting the initial model according to the optimal model order to obtain a multiple autoregressive model;
and 4, step 4: under the transient effect when the lag variable is equal to 0, carrying out Fourier transform on the multiple autoregressive model, converting the multiple autoregressive model into a frequency domain and extracting a model coefficient matrix;
and 5: calculating the extended biased directional coherent values between all channel pairs according to the model coefficient matrix extracted in the step 4;
step 6: verifying the statistical significance of the extended biased directional coherence value between each pair of the cerebral muscle channels in a frequency domain by adopting a causal Fourier transform proxy data method, and correcting the extended biased directional coherence value between the corresponding pair of the cerebral muscle channels to be 0 if the extended biased directional coherence value does not have the statistical significance;
and 7: quantitatively analyzing the obviously expanded biased directional coherent area between channels In a specific frequency band as functional connectivity measurement, establishing a correlation matrix, and respectively forming two complex brain-muscle interaction closed-loop functional networks according to a defined Out direction (Out direction) and In direction (In direction). The out-degree direction and the in-degree direction are specifically as follows: the different information flows between all pairs of the brain muscle channels are defined as the outbound direction and the inbound direction, respectively, according to two regions divided by the diagonal of the correlation matrix.
Further, in step 6, a causal fourier transform proxy data method is used to verify the statistical significance of the extended biased directional coherence value between each pair of brain muscle channels in the frequency domain, which is specifically as follows:
step 6.1: by randomizing the amplitude and phase angle of the original N + M dimensional time sequence, the direct causal coupling is destroyed, but the power spectrum and phase difference are preserved;
step 6.2: and determining a confidence interval and a significance threshold value by adopting a renormalized biased directional coherent method. The significance threshold is defined as the 95 th percentile of each coherence distribution in the 1000 agent sets; and verifying the statistical significance of the extended biased directional coherence value between each pair of the brain muscle channels in the frequency domain according to a significance threshold value.
Further, in step 7, a significantly extended partially oriented coherent area between channels in a specific frequency band is quantitatively analyzed as a functional connectivity metric, which is specifically as follows:
step 7.1: constructing a functional connectivity framework related to related events among different cortical areas, cortical muscles and different muscles;
step 7.2: analysing specific frequency bands (f)b～fe) The significant coherence area between the middle channels is normalized to a functional connectivity metric. Wherein the normalized significant spread between two channels is biased to a directional coherent region FCePDCQuantification is performed by the following formula:
Figure BDA0003520415100000022
Figure BDA0003520415100000023
representing the spread-out biased directional coherence value, SL, at frequency fij(f) Denotes the significance threshold at frequency f, Δ f denotes the frequency resolution, r denotes different source channel combinations, r =1,2,3,rijRepresenting multiple autoregressive variables.
Further, the method also comprises the effective connectivity analysis of the closed-loop network, specifically comprising the following steps: the total coherence information flowing from a source region to a destination region is defined as an inflow index (IFi) indicating the causal impact of one region on another region of the closed loop system, and the total coherence information flowing from the destination region to the source region is defined as an outflow index (OFi) indicating the extent to which one region is causally driven by another region. Respectively, as follows:
Figure BDA0003520415100000031
Figure BDA0003520415100000032
Figure BDA0003520415100000033
indicating a specific frequency band fb～feThe offset between the two channels introduces coherent flow information,
Figure BDA0003520415100000034
indicating a biased directional coherent outflow of information between the two channels in a particular frequency band.
Further, in step 7, the correlation matrix is converted into a weighted graph by using the network sparsity, i.e., the connection density, as a threshold measurement, so as to obtain two complex brain-muscle interaction closed-loop functional networks.
Further, the dynamic change of the cortex-muscle-cortex coordination work mode is characterized by using the edge strength of the closed-loop network.
The invention has the beneficial effects that: the invention provides an information coupling calculation frame with strong noise robustness under low delay signals (simulating the information transmission process of a brain neuron group), and meanwhile, the elasticity of synchronous myoelectric information is kept; the invention designs a communicating index which is more in line with physiological characteristics, namely the quantitative measurement of asymmetric coupling information, and estimates the interaction strength between brain functional areas and between brain-muscle coupling areas in a small frequency band. The invention provides a complex brain-muscle interaction closed-loop function network, which can comprehensively explain the activity change of a patient cortex and muscle lesion part at an electrophysiological level, and the system view allows the pathophysiology of the movement disorder caused by the stroke to have top-down (brain-muscle) topological property to be understood, thereby helping a follow-up doctor to make an intervention treatment strategy.
Drawings
FIG. 1 is a schematic diagram of the overall and exploded view of a preferred embodiment of the present invention;
FIG. 2 is a conceptual diagram of a complex brain-muscle interaction closed loop functional network according to a preferred embodiment of the present invention.
Fig. 3 is a schematic diagram of the functional connectivity topology of a preferred embodiment of the present invention.
FIG. 4 is a schematic diagram of an efficient connectivity analysis topology according to a preferred embodiment of the present invention.
FIG. 5 is a schematic diagram of a correlation matrix and a weighted closed-loop network according to a preferred embodiment of the present invention.
Detailed Description
The technical contents of the preferred embodiments of the present invention will be more clearly and easily understood by referring to the drawings attached to the specification. The present invention may be embodied in many different forms of embodiments and the scope of the invention is not limited to the embodiments set forth herein.
The complex brain-muscle interaction closed-loop functional network proposed by the invention is realized according to the method steps shown in figure 1. The existing general method for connectivity (functional connectivity) analysis in brain network research based on scalp brain electrical activity is mainly based on phase, amplitude or combined amplitude-phase measurements. Multiple sensors register volume conduction or field diffusion resulting from source activity, leading to the presence of spurious linear correlations that are highly likely to negatively impact the estimation of statistical correlations between time series. The invention proposes a statistical measure which is insensitive to volume conduction and which can effectively characterize the coupling strength between the brain and the muscle, in order to avoid overestimating the connectivity due to false interactions, comprising the following steps:
step 1: combining the collected and recorded electroencephalogram (EEG) and Electromyogram (EMG) data into an N + M-dimensional time sequence; wherein, N is the channel number of the electroencephalogram data, and M is the channel number of the electromyogram data.
Step 2: the method is expanded by adopting a multivariate causal estimation method insensitive to the volume conduction effect to weaken the volume conduction effect, and specifically comprises the following steps: the above causal time series is described as a multiple autoregressive process, expressed as follows:
Figure BDA0003520415100000041
wherein Xr(t) is a time series, AkFor sparse matrices, p represents the model order, k represents the lag variable, εr(t) representing white noise, and expanding the white noise into a transient effect and a hysteresis effect by setting a value of 0 or non-0 according to a hysteresis variable k to obtain an initial model; r denotes different source channel combinations, r =1,2,3.
And step 3: estimating the optimal model order of the initial model based on Akaike information standard, and correcting the initial model according to the optimal model order to obtain a multiple autoregressive model; to avoid overfitting, the maximum order is typically limited to 20;
and 4, step 4: further converting the correlation between residuals into model coefficients to fully describe the direct interaction between time series, i.e. between channels, specifically: under the transient effect when the lag variable is equal to 0, carrying out Fourier transform on the multiple autoregressive model, converting the multiple autoregressive model into a frequency domain and extracting a model coefficient matrix;
and 5: the method comprises the following steps of constructing an asymmetric biased directional coherence measure of functional connection under a frequency domain by combining an extended multivariate autoregressive (MVAR) model with instantaneous and hysteresis effects, and realizing effective estimation of a frequency domain causal relationship in a narrow-band frequency range, wherein the method specifically comprises the following steps: calculating an extended bias directional coherent value between the pair of the brain muscle channels according to the model coefficient matrix extracted in the step 4 to form an initial correlation matrix between the pair of the brain muscle channels;
and 6: and (3) verifying the statistical significance of the extended biased directional coherence value between each pair of the brain muscle channels in a frequency domain by adopting a causal Fourier transform proxy data method, which comprises the following steps:
step 6.1: by randomizing the amplitude and phase angle of the original N + M dimensional time sequence, the direct causal coupling is destroyed, but the power spectrum and phase difference are preserved;
step 6.2: and determining a confidence interval and a significance threshold value by adopting a renormalized biased directional coherent method. The significance threshold is defined as the 95 th percentile of each coherence distribution in the 1000 agent sets;
if the statistical significance does not exist, correcting the expansion deviation orientation coherent value between the corresponding brain muscle channel pair to be 0 to form a corrected correlation matrix; the corrected asymmetric biased directional coherence metric in the correlation matrix can evaluate hysteresis effects and causal relationships in an extended form including transient effects, and is suitable for functional connection analysis between different brain regions with a large number of transient effects and brain muscle with hysteresis effects.
And 7: quantitatively analyzing the significantly extended biased directional coherent area between channels in a specific frequency band as a functional connectivity metric, specifically as follows:
step 7.1: constructing a functional connectivity framework involving events related between different cortical areas, cortical muscles and different muscles;
step 7.2: analysing specific frequency bands (f)b～fe) The significant coherence area between the middle channels is normalized to a functional connectivity metric. Wherein the normalized significant spread between two channels is biased directional coherent region FCePDCQuantification is performed by the following formula:
Figure BDA0003520415100000051
Figure BDA0003520415100000052
representing the spread-out partially directional coherence value, SL, at frequency fij(f) Denotes the significance threshold at frequency f, af denotes the frequency resolution, r denotes different source channel combinations, r =1,2,3 (brain-brain channel pair, brain-muscle channel pair, muscle-muscle channel pair, respectively), rijRepresenting multiple autoregressive variables.
According to two areas divided by the diagonal line of the coherence matrix, different information flows between all brain muscle channel pairs are respectively defined as Out direction and In direction; as shown in FIG. 5, in this embodiment, the lower left corner is defined as the out-degree direction and the upper right corner is defined as the in-degree direction, where channels 1-32 and channels 33-40 represent the EEG and EMG, respectively. Taking the output direction as an example, fig. 3 shows a functional connectivity topology based on a complex brain-muscle interaction closed-loop functional network, wherein solid lines, dashed lines and dotted lines show non-zero metric values between different channels.
A schematic of a complex brain-muscle interaction closed loop functional network is shown in fig. 2. From left to right are the inter-channel interaction and the multi-channel interaction, respectively. Different styles of curves represent different interactive information flows (out-degree and in-degree).
Further, in principle, the network can be abstractly represented as a graph consisting of vertices and corresponding edges. According to the method, the edge strength of a closed-loop network is considered, and constraint least square weighted summation is carried out on the asymmetric biased directional coherence measure. The method comprises the following specific steps:
converting the brain muscle interaction connection problem into a weighted graph based on the optimized weighted value;
further establishing a standardized weighted value as a connecting edge of the brain-muscle closed loop function network, and converting each correlation matrix into a weighted graph by taking a network sparsity function, namely, a connection density as a threshold value;
and setting the weighted edge value lower than the threshold value as 0, reducing the network complexity to improve the local significance, and finally generating the weighted brain-muscle closed-loop functional network. The weighting network for the information egress and ingress directions is shown on the right side of fig. 5. And visually representing the dynamic change of the cortex-muscle-cortex coordination work mode through the edge strength visualization of the closed-loop network.
The invention provides a complex brain-muscle interaction closed-loop function network by means of multi-channel brain electromyogram signals, an informatization measure is fused into the brain function network in an interactive connection mode, a complete closed-loop network system of brain function network and brain muscle coupling and inter-muscle coupling interaction is constructed, the whole brain area and local damaged muscle tissues are covered, the activity change of a patient cortex and a muscle pathological change part is comprehensively explained at an electrophysiological level, the system visual angle allows the understanding of the topological property from top to bottom (from brain to muscle) on the pathophysiology of dyskinesia caused by stroke, and the formulation of a follow-up doctor intervention treatment strategy is facilitated.
Preferably, in order to maintain the elasticity of the synchronous noise and the electroencephalogram and electromyography information, the multivariate autoregressive model constructed in the step 3 can be further modified by data modeled by neuron signal simulation, and can be used for example
Figure BDA0003520415100000053
The oscillator models the neuron population and generates data sets with different noise synchronization levels, the time series coupling intensity is between 0 and 1, and the evolution of the phase difference follows a linear trend, namely in the range of [ -pi, pi]Converting the completely equal frequency component with constant phase difference into differential phase within the range; when the frequency of the signal is shifted, the method verifies the coupling detection capability of the causal relationship model and the robustness of noise based on the EEG and EMG data sets, and establishes the optimal model coefficient with strong robustness.
As another preference, effective connectivity represents the effect one nervous system (or brain region) exerts on another nervous system, and can be quantified by various interaction or causal coupling models, such as the guillain causal relationship and dynamic causal models. Given that biased directional coherence is a frequency domain representation of the granger causal relationship, closed loop functional networks can be characterized to some extent by effective connectivity.
An effective connectivity analysis of the closed-loop functional network is performed based on the assumption that the brain and muscle tissue regions are characterized and that a region of interest (ROI) is defined. Brain ROIs extraction followed the following criteria: ROIs are partitioned by brain atlas guidance, with no overlapping regions, ensuring the selection of functionally meaningful connected regions. In particular, for brain regions, the present invention employs cortical reconstruction of a head model as a realistic geometric head model of a subject. The electrical brain signals acquired from the standard electrode leads of the international 10/20 system are represented on a true geometric model.
FIG. 4 shows a topology for efficient connectivity analysisI.e. the overall effect of different ROIs on the information flow, the arrow direction represents the IFi and OFi parameters. Among the 8 ROIs, the motor cortex area of the brain on the healthy and affected sides was defined as MuAnd Ma(ii) a The sensory cortex of healthy and affected sides is defined as SuAnd Sa(ii) a The associated area of the apical lobe between healthy and affected sides is defined as PAuAnd PAa. Aiming at muscle areas related to two anatomical parts of an upper limb, according to signals collected by a myoelectric electrode at an affected side, proximal muscles (biceps brachii, triceps brachii and deltoid muscle) and distal muscles (superficial flexor digitorum, flexor ulnaris, flexor radialis, extensor carpi ulnaris and extensor radialis longus) of the upper limb are divided into two ROIs which are respectively defined as PM and DM.
The above description is only a preferred embodiment of the present invention, and it should be understood that several modifications and variations can be made in the concept of the present invention by those skilled in the art without inventive step. Therefore, the technical solutions that can be obtained by a person skilled in the art through logical reasoning or experiments based on the concepts of the present invention in the prior art should be considered as the protection scope of the present invention.

Claims (4)

1. A method for constructing a complex brain-muscle interaction closed-loop function network framework is characterized by comprising the following steps:
step 1: combining the acquired and recorded electroencephalogram data and electromyogram data into an N + M-dimensional time sequence; wherein, N is the channel number of the electroencephalogram data, and M is the channel number of the electromyogram data;
step 2: describing the time series as a multiple autoregressive process, as follows:
Figure FDA0003845190370000011
wherein Xr(t) is a time series, AkFor sparse matrices, p represents the model order, k represents the lag variable, εr(t) representing white noise, and expanding the white noise into a transient effect and a hysteresis effect by setting a value of 0 or non-0 according to a hysteresis variable k to obtain an initial model;
and step 3: estimating the optimal model order of the initial model based on Akaike information standard, and correcting the initial model according to the optimal model order to obtain a multiple autoregressive model;
and 4, step 4: under the transient effect when the lag variable is equal to 0, carrying out Fourier transform on the multiple autoregressive model, converting the multiple autoregressive model into a frequency domain and extracting a model coefficient matrix;
and 5: calculating the extended bias directional coherent values between all channel pairs according to the model coefficient matrix extracted in the step 4;
step 6: and (3) verifying the statistical significance of the extended biased directional coherence value between each pair of the brain muscle channels in a frequency domain by adopting a causal Fourier transform proxy data method, which comprises the following steps:
step 6.1: by randomizing the amplitude and phase angle of the original N + M dimensional time sequence, the direct causal coupling is destroyed, but the power spectrum and phase difference are preserved;
step 6.2: determining a confidence interval and a significance threshold value by adopting a renormalized partial directional coherent method; the significance threshold is defined as the 95 th percentile of each coherence distribution in the 1000 agent sets; verifying the statistical significance of the extended biased directional coherence value between each pair of the brain muscle channels in a frequency domain according to a significance threshold; if the statistical significance does not exist, correcting the expansion deviation orientation coherent value between the corresponding brain muscle channel pair to be 0;
and 7: quantitatively analyzing a remarkably expanded biased directional coherent area between channels in a specific frequency band as functional connectivity measurement, establishing a correlation matrix, and respectively forming two complex brain-muscle interaction closed-loop functional networks according to a defined out-degree direction and an in-degree direction; the out-degree direction and the in-degree direction are specifically as follows: according to two areas divided by the diagonal line of the correlation matrix, different information flows between all the brain muscle channel pairs are respectively defined as out-degree directions and in-degree directions;
the specific measurement of functional connectivity is as follows by quantitatively analyzing the significantly extended partially oriented coherent area between channels in a specific frequency band:
step 7.1: constructing a functional connectivity framework related to related events among different cortical areas, cortical muscles and different muscles;
step 7.2: analysing specific frequency bands fb～feNormalizing and quantifying the obvious coherent area between the middle channels into a functional connectivity metric value; wherein the normalized significant spread between two channels is biased to a directional coherent region FCePDCQuantification is performed by the following formula:
Figure FDA0003845190370000021
Figure FDA0003845190370000022
representing the spread-out partially directional coherence value, SL, at frequency fij(f) Denotes a significance threshold at frequency f, Δ f denotes the frequency resolution, r denotes different source channel combinations, r =1,2,3,rijRepresenting multiple autoregressive variables.
2. The method according to claim 1, further comprising a closed-loop network effective connectivity analysis, specifically: defining the total coherent information flowing from the source region to the target region as an inflow index IFi representing the causal effect of one region on another region of the closed-loop system, and defining the total coherent information flowing from the target region to the source region as an outflow index OFi representing the extent to which one region is causally driven by another region; respectively, as follows:
Figure FDA0003845190370000023
Figure FDA0003845190370000024
Figure FDA0003845190370000025
to indicate a particularFrequency band fb～feThe offset between the two channels introduces coherent flow information,
Figure FDA0003845190370000026
indicating a biased directional coherent outflow of information between two channels in a particular frequency band.
3. The construction method according to any one of claims 1-2, wherein in step 7, two complex brain-muscle interaction closed-loop function networks are obtained by converting the correlation matrix into a weighted graph using network sparsity (i.e. connection density) as a threshold measure.
4. The method of construction according to any one of claims 1-2, wherein the dynamic changes of the cortical-muscle-cortical coordination mode of operation are characterized by using the edge strength of the closed-loop network.
CN202210176343.8A 2022-02-25 2022-02-25 Method for constructing complex brain muscle interaction closed loop function network framework Active CN114569139B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210176343.8A CN114569139B (en) 2022-02-25 2022-02-25 Method for constructing complex brain muscle interaction closed loop function network framework

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210176343.8A CN114569139B (en) 2022-02-25 2022-02-25 Method for constructing complex brain muscle interaction closed loop function network framework

Publications (2)

Publication Number Publication Date
CN114569139A CN114569139A (en) 2022-06-03
CN114569139B true CN114569139B (en) 2022-11-01

Family

ID=81773435

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210176343.8A Active CN114569139B (en) 2022-02-25 2022-02-25 Method for constructing complex brain muscle interaction closed loop function network framework

Country Status (1)

Country Link
CN (1) CN114569139B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102488515A (en) * 2011-12-09 2012-06-13 天津大学 Conjoint analysis method for electroencephalograph and electromyography signals based on autonomous movement and imagination movement
CN105595961A (en) * 2015-12-21 2016-05-25 天津大学 Alzheimer's disease detecting system and method based on electroencephalogram signals
CN107193374A (en) * 2017-05-17 2017-09-22 中国兵器工业计算机应用技术研究所 A kind of detection means and detection method of the intentional gesture motion of active
CN112545519A (en) * 2021-02-22 2021-03-26 之江实验室 Real-time assessment method and system for group emotion homogeneity
CN113827249A (en) * 2021-09-28 2021-12-24 之江实验室 Method for eliminating multichannel gain error of electroencephalogram signal acquisition system

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090062680A1 (en) * 2007-09-04 2009-03-05 Brain Train Artifact detection and correction system for electroencephalograph neurofeedback training methodology
CN103942424B (en) * 2014-04-13 2017-02-08 北京师范大学 Large-scale cortex network information flow individual three-dimensional dynamic visualization method
US11717686B2 (en) * 2017-12-04 2023-08-08 Neuroenhancement Lab, LLC Method and apparatus for neuroenhancement to facilitate learning and performance
WO2020107034A1 (en) * 2018-11-23 2020-05-28 Wellness Ip, Inc. Micro-coherence network strength and deep behavior modification optimization application
CN112617858A (en) * 2020-12-01 2021-04-09 杭州电子科技大学 Cortical muscle function coupling method for multi-scale compensation transfer entropy

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102488515A (en) * 2011-12-09 2012-06-13 天津大学 Conjoint analysis method for electroencephalograph and electromyography signals based on autonomous movement and imagination movement
CN105595961A (en) * 2015-12-21 2016-05-25 天津大学 Alzheimer's disease detecting system and method based on electroencephalogram signals
CN107193374A (en) * 2017-05-17 2017-09-22 中国兵器工业计算机应用技术研究所 A kind of detection means and detection method of the intentional gesture motion of active
CN112545519A (en) * 2021-02-22 2021-03-26 之江实验室 Real-time assessment method and system for group emotion homogeneity
CN113827249A (en) * 2021-09-28 2021-12-24 之江实验室 Method for eliminating multichannel gain error of electroencephalogram signal acquisition system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Corticomuscular interactions during different movement periods in a multi-joint compound movement;Rouven Kenville et al;《Scientific Reports》;20201231;全文 *
基于MEMD-rTVgPDC的皮层肌肉耦合分析;张敏 等;《传感技术学报》;20200531;第33卷(第3期);全文 *

Also Published As

Publication number Publication date
CN114569139A (en) 2022-06-03

Similar Documents

Publication Publication Date Title
CN102573619B (en) Device and method for generating a representation of a subject's attention level
CN110495880B (en) Movement disorder cortical plasticity management method based on transcranial electrical stimulation brain muscle coupling
Yang et al. Nonlinear coupling between cortical oscillations and muscle activity during isotonic wrist flexion
CN108309318B (en) Brain functional state evaluation device based on brain hemoglobin information
Moran et al. Revealing neuronal functional organization through the relation between multi-scale oscillatory extracellular signals
CN109640810B (en) Method of modulating epileptogenesis in the brain of a patient
CN110013250B (en) Multi-mode characteristic information fusion prediction method for suicidal behavior of depression
CN111096745B (en) Steady-state evoked response brain source positioning method based on sparse Bayesian learning
Ranaldi et al. An automatic, adaptive, information-based algorithm for the extraction of the sEMG envelope
Rohlén et al. A method for identification of mechanical response of motor units in skeletal muscle voluntary contractions using ultrafast ultrasound imaging—simulations and experimental tests
Porcaro et al. Choice of multivariate autoregressive model order affecting real network functional connectivity estimate
CN112957014A (en) Pain detection and positioning method and system based on brain waves and neural network
Zavaglia et al. The effect of connectivity on EEG rhythms, power spectral density and coherence among coupled neural populations: analysis with a neural mass model
CN104887198A (en) Pain quantitative analysis system and method based on human body physiological signal multi-parameter fusion
Chen et al. An IC-PLS framework for group corticomuscular coupling analysis
CN114027857A (en) Method for measuring kinetic energy based on electroencephalogram signals
CN114569139B (en) Method for constructing complex brain muscle interaction closed loop function network framework
Parthiban et al. Spike detection from electroencephalogram signals with aid of hybrid genetic algorithm-particle swarm optimization
CN113440150A (en) Cortical muscle function network construction method based on R-vine Copula
Shi et al. Effect of the small-world structure on encoding performance in the primary visual cortex: an electrophysiological and modeling analysis
Falcon-Caro et al. A diffusion adaptation approach to model brain responses in an EEG-based hyperscanning study
He et al. Measuring temporal dynamics of resting-state fMRI data
GÜl et al. Automated pre-seizure detection for epileptic patients using machine learning methods
CN113269707A (en) fMRI data function connection method based on work memory task and application thereof
Severini et al. A multivariate auto-regressive method to estimate cortico-muscular coherence for the detection of movement intent

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