WO2021046949A1 - 一种驾驶疲劳相关的eeg功能连接动态特性的分析方法 - Google Patents

一种驾驶疲劳相关的eeg功能连接动态特性的分析方法 Download PDF

Info

Publication number
WO2021046949A1
WO2021046949A1 PCT/CN2019/109853 CN2019109853W WO2021046949A1 WO 2021046949 A1 WO2021046949 A1 WO 2021046949A1 CN 2019109853 W CN2019109853 W CN 2019109853W WO 2021046949 A1 WO2021046949 A1 WO 2021046949A1
Authority
WO
WIPO (PCT)
Prior art keywords
time
eeg
data
temporal
brain network
Prior art date
Application number
PCT/CN2019/109853
Other languages
English (en)
French (fr)
Inventor
王洪涛
刘旭程
李霆
唐聪
裴子安
岳洪伟
陈鹏
许弢
Original Assignee
五邑大学
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 五邑大学 filed Critical 五邑大学
Priority to US16/975,683 priority Critical patent/US20210267513A1/en
Priority to AU2019433218A priority patent/AU2019433218A1/en
Publication of WO2021046949A1 publication Critical patent/WO2021046949A1/zh

Links

Images

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/16Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
    • A61B5/18Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state for vehicle drivers or machine operators
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/16Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
    • A61B5/165Evaluating the state of mind, e.g. depression, anxiety
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/16Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
    • A61B5/168Evaluating attention deficit, hyperactivity
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316Modalities, i.e. specific diagnostic methods
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316Modalities, i.e. specific diagnostic methods
    • A61B5/369Electroencephalography [EEG]
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/72Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7203Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/72Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235Details of waveform analysis
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/72Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235Details of waveform analysis
    • A61B5/7253Details of waveform analysis characterised by using transforms
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/10Pre-processing; Data cleansing
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/21Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/213Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
    • G06F18/2134Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on separation criteria, e.g. independent component analysis
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B2503/00Evaluating a particular growth phase or type of persons or animals
    • A61B2503/20Workers
    • A61B2503/22Motor vehicles operators, e.g. drivers, pilots, captains
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316Modalities, i.e. specific diagnostic methods
    • A61B5/369Electroencephalography [EEG]
    • A61B5/372Analysis of electroencephalograms
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316Modalities, i.e. specific diagnostic methods
    • A61B5/369Electroencephalography [EEG]
    • A61B5/372Analysis of electroencephalograms
    • A61B5/374Detecting the frequency distribution of signals, e.g. detecting delta, theta, alpha, beta or gamma waves
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316Modalities, i.e. specific diagnostic methods
    • A61B5/398Electrooculography [EOG], e.g. detecting nystagmus; Electroretinography [ERG]
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/72Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235Details of waveform analysis
    • A61B5/725Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/72Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235Details of waveform analysis
    • A61B5/7253Details of waveform analysis characterised by using transforms
    • A61B5/726Details of waveform analysis characterised by using transforms using Wavelet transforms

Definitions

  • the invention relates to the technical field of driving fatigue analysis, in particular to a method for analyzing the dynamic characteristics of EEG functional connections related to driving fatigue.
  • FC fatigue-related EEG functional connection
  • the purpose of the present invention is to provide a method for analyzing the dynamic characteristics of driving fatigue-related EEG functional connections, and applying the dynamic FC analysis framework to the driving fatigue research, so as to obtain driving fatigue-related brains on a fine time scale.
  • a method for analyzing the dynamic characteristics of EEG functional connections related to driving fatigue including:
  • the correlation includes spatio-temporal global efficiency, spatio-temporal local efficiency, and temporal-spatial proximity centers degree.
  • preprocessing of EEG data using independent component analysis and wavelet packet transform includes:
  • the blinking artifact data includes horizontal EOG HEOG data and vertical EOG VEOG data;
  • the EEG data is divided into awake state data and fatigue state data.
  • the construction of the preprocessed EEG data into a time brain network with dynamic characteristics based on the sliding window method includes:
  • the static network being a binary NxN matrix, where N represents the number of electrodes of the EEG cap;
  • the PLI value above the threshold is set to 1, and the PLI value below the threshold is set to 0, thereby forming a binary adjacency network as a snapshot of the temporal brain network;
  • the static network is arranged according to the time sequence to form a time-brain network with dynamic characteristics.
  • the measuring the spatiotemporal topology of the temporal brain network based on the time efficiency analysis framework includes:
  • the temporal and spatial proximity centrality is used to evaluate the temporal and spatial characteristics of the node-level temporal brain network.
  • An analysis device for the dynamic characteristics of the EEG functional connection related to driving fatigue including:
  • the preprocessing unit is used to preprocess the EEG data using independent component analysis and wavelet packet transform
  • the construction unit is used to construct the preprocessed EEG data into a dynamic time brain network based on the sliding window method
  • a spatiotemporal topology calculation unit configured to measure the spatiotemporal topology of the time brain network based on a time efficiency analysis framework
  • the statistical analysis unit is used to perform statistical analysis on the spatio-temporal topology of the time-brain network to obtain the correlation between the driving fatigue-related behavior performance and the dynamic characteristics of the time-brain network.
  • the correlation includes the spatio-temporal global efficiency and the spatio-temporal Local efficiency and temporal proximity to centrality.
  • the pre-processing unit includes:
  • a blink artifact data collection unit for the blink artifact data including horizontal electro-oculogram HEOG data and vertical electro-oculogram VEOG data;
  • a filtering unit configured to use independent component analysis to find and delete components in the EEG data that are highly related to the blinking artifact data
  • the baseline removal unit is used to remove the baseline of the filtered EEG data
  • a decomposition unit configured to use wavelet packet transform to decompose the EEG data into three standard frequency bands, which are alpha frequency band, beta frequency band and ⁇ frequency band;
  • the data division unit is used to divide the EEG data into awake state data and fatigue state data according to the test time.
  • construction unit includes:
  • the sliding window processing unit is used to select a suitable window length and step length, the sliding window sequentially traverses the entire time series, and the length of the time series is the length of the experiment for collecting EEG data;
  • the PLI calculation unit is configured to use the phase lag index to estimate the PLI value of the functional connection in each of the static networks;
  • the binarization calculation unit adopts the sparsity method to set the PLI value higher than the threshold to 1, and the PLI value lower than the threshold to 0, thereby forming a binarized adjacency network, which is used as the time Snapshot of the brain network;
  • the sparsity calculation unit is used to select an appropriate sparsity and interval, and use sparsity methods to reserve the required functional connections in each static network;
  • the dynamic characteristic building unit is used for arranging the static network according to the time sequence to form a time brain network with dynamic characteristics.
  • spatiotemporal topology calculation unit includes:
  • a time distance calculation unit for calculating the time distance of a pair of nodes on a time scale, where the time distance represents the minimum number of time windows defined as a time-space path passed;
  • the space-time global efficiency calculation unit is used to calculate the space-time global efficiency
  • the space-time local efficiency calculation unit is used to calculate the space-time local efficiency
  • the spatiotemporal feature evaluation unit is used to evaluate the spatiotemporal features of the node-level temporal brain network using the temporal proximity centrality.
  • a device for analyzing the dynamic characteristics of the EEG functional connection related to driving fatigue comprising at least one control processor and a memory for communicating with the at least one control processor; the memory stores the memory that can be processed by the at least one control processor.
  • the instructions executed by the driver, the instructions are executed by the at least one control processor, so that the at least one control processor can execute the method for analyzing the dynamic characteristics of the EEG functional connection related to driving fatigue as described in any one of the above.
  • a computer-readable storage medium characterized in that: the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to make a computer execute the driving fatigue-related EEG described in any one of the above The analysis method of the dynamic characteristics of the functional connection.
  • the one or more technical solutions provided in the embodiments of the present invention have at least the following beneficial effects: by introducing time characteristics into the static network of driving fatigue, constructing a time brain network with dynamic characteristics, and obtaining the time during driving fatigue through analysis and statistics.
  • the analysis method of the present invention has more accurate analysis results, which is beneficial to reveal the reorganization of the information transmission function between the brain regions related to driving fatigue on the fine time scale. More critical dynamic characteristics.
  • Figure 1 is a flow chart of the overall method of an embodiment of the present invention.
  • Figure 2 is a flowchart of a preprocessing method according to an embodiment of the present invention.
  • FIG. 3 is a flowchart of constructing a time-brain network with dynamic characteristics according to an embodiment of the present invention
  • FIG. 4 is a flowchart of measuring the spatio-temporal topology of a time brain network according to an embodiment of the present invention
  • FIG. 5 is a schematic diagram of a unit architecture in an apparatus according to an embodiment of the present invention.
  • Figure 6 is a schematic diagram of connections in a device of an embodiment of the present invention.
  • an embodiment of the present invention provides a method for analyzing the dynamic characteristics of EEG functional connections related to driving fatigue, including:
  • S4 Perform statistical analysis on the spatio-temporal topology of the time-brain network to obtain the correlation between the driving fatigue-related behavior performance and the dynamic characteristics of the time-brain network.
  • the correlation includes spatio-temporal global efficiency, spatio-temporal local efficiency, and spatio-temporal Proximity to centrality.
  • step S1 includes:
  • S14 Use wavelet packet transform to decompose the EEG data into three standard frequency bands, which are alpha frequency band, beta frequency band, and ⁇ frequency band;
  • S15 Divide the EEG data into awake state data and fatigue state data according to the test time.
  • step S2 includes:
  • S21 representing the preprocessed EEG data as a static network, the static network being a binary NxN matrix, where N represents the number of electrodes of the EEG cap;
  • the binarization calculation unit adopts the sparsity method to set the PLI value higher than the threshold to 1, and the PLI value lower than the threshold to 0, so as to form a binarized adjacency network, which is used as the total value.
  • S25 Arranging the static network according to the time sequence to form a time-brain network with dynamic characteristics.
  • step S3 includes:
  • S31 Calculate the time distance of the pair of nodes on a time scale, where the time distance represents a minimum number of time windows defined as a time-space path passed;
  • connection (FC) in driving fatigue research is static, that is, a representative brain network is constructed with a time scale of a few minutes in a fatigue state. Since a key component of driving fatigue is task time itself, it is emphasized that the brain regulates fatigue. Because of the accumulation characteristics of the process of performance degradation, static brain network research lacks more critical dynamic characteristics about the reorganization of information transmission functions between brain regions related to driving fatigue on fine time scales.
  • the experimental setting of the embodiment of the present invention is that the subject conducts simulated driving for 90 minutes, and a virtual guided vehicle is set up in front of the subject.
  • the virtual guided vehicle generates braking signals at random intervals and requires the subject to pass the system. Respond to the braking signal to maintain a safe distance.
  • the time interval from the braking command generated in the virtual guided vehicle to the subject's braking operation is considered as the reaction time (RT), and the vehicle's speed change (SV) is also collected as a quantitative evaluation of the subject's behavior index.
  • RT reaction time
  • SV vehicle's speed change
  • the data collection method is that the subject wears a 24-channel wireless EEG cap with an improved international 10-20 electrode placement system (HD-72, Cognionics, Inc., USA) to record EEG data at 250 Hz.
  • the reference electrodes are right and left.
  • the EEG signal is filtered by a band-pass filter (between 2 and 100 Hz).
  • blinking artifacts were recorded by electrodes placed on the outer corner of the eye (horizontal electro-oculogram, HEOG) and above and below the right eye (vertical electro-oculogram, VEOG).
  • ICA Use independent component analysis
  • the dynamic analysis framework of brain network is to use the optimal sliding window to characterize the time-ordered static network.
  • the phase lag index (PLI) is used to estimate FC, because it has advantages in minimizing the influence of common source signal and volume conduction.
  • the PLI value of FC is between 0 and 1, and the larger the number, the larger the value. The stronger the connection.
  • using the commonly used sparsity method select a sparsity of 5%-15%, with an interval of 1%, and threshold the matrix into a binarized adjacency network, which is equivalent to comparing the PLI value of FC with the threshold, and the setting is greater than the threshold.
  • the choice of window length can be 3-6 seconds, and the choice of step size can be 2-4 seconds.
  • the window length is chosen to be 4 Second, the step length is selected as 4 seconds.
  • the most awake and fatigued state are determined by statistical comparison of the subjects’ behavioral performance in terms of reaction time and speed change.
  • the spatio-temporal topology of the time brain network is the spatio-temporal topology of the time brain network
  • the time distance is defined as the minimum number of time windows passed by the time-space path. It is worth noting that the time-dependent path is Metrics in the space and time domains, and the time distance is represented by the time domain. Therefore, the time distance is a positive integer, and its range is between 1 and T.
  • the spatiotemporal efficiency analysis framework is used to measure the spatiotemporal topology of the time brain network.
  • the spatiotemporal efficiency measures the interaction and information transmission functions between the entire nodes in the dynamic system, and the spatiotemporal global efficiency The ability to capture the dynamics of the entire network and the information flow is spread throughout the entire life cycle. The calculation method is as follows:
  • G is a spatiotemporal network with a mathematical structure NxNxT, and It is the efficiency of space-time global efficiency at time t, and the value of t is a positive integer not greater than T.
  • Ranges from 0 to 1 It means that all nodes are connected in the snapshot of the spatiotemporal brain network, ⁇ i ⁇ j (t) means the time distance from i to j at time t. .
  • the temporal and spatial local efficiency measures the overall elasticity of the dynamic network to randomly remove nodes in a local range:
  • G(i,t) is a sub-spatiotemporal network including all neighbors of node i at time t.
  • the range is from 0 to 1. It is an index of a dynamic network, used to measure the information dissemination capacity of a local scale.
  • the temporal and spatial proximity centrality is used to evaluate the temporal and spatial characteristics of the node-level temporal brain network.
  • the centrality measures the ability of node i to reach other nodes C c (i, t) is expressed as:
  • the temporal and spatial proximity centrality also indicates the importance of nodes in the entire temporal network.
  • the integral space-time proximity centrality represents the area under the curve of node i in the entire sparsity range.
  • the reference network increases the randomness of the dynamic brain network, showing different information dissemination efficiencies. Calculating such a reference network helps to reveal the neural mechanism of dynamic FC with different topologies.
  • a two-step randomization method is used: random edge (RE) and random connection (RC).
  • the RE method randomly reconnects all edges in the dynamic network under certain constraints, which destroys the topology of the dynamic brain network.
  • the RC method randomly redistributes all connections in the network, which eliminates the distribution of the number of connections at each edge.
  • the application of these two methods destroys the main structure of the dynamic network.
  • the definition of the small-world attribute in the static network is extended to the time-brain network with dynamic characteristics. If the following definition is satisfied, the time-brain network with dynamic characteristics is small-world in space and time:
  • the spatiotemporal efficiency of the reference network for each subject is the average of the generated spatiotemporal reference network with 50 iterations in two mental states (awake and fatigue).
  • FDR False Discovery Rate
  • reaction time (RT) and speed change (SV) in the 5 minutes before and after are respectively expressed in the form of coordinates to form a time-response time coordinate graph and a time-speed change coordinate graph, which can be clearly seen in the first 5 minutes There is a significant difference between the two states and the last 5 minutes, which confirms that the first 5 minutes are awake and the last 5 minutes are fatigued.
  • the spatiotemporal topology of brain activity is quantitatively estimated by the spatiotemporal efficiency of the awake and fatigue states, and the frequency band-efficiency graph is established according to the three standard frequency bands to show the integrated spatiotemporal global efficiency and integrated spatiotemporal localization in the entire sparse range
  • the time-brain network with efficiency and dynamic characteristics exhibits an outstanding spatio-temporal small-world architecture in all three frequency bands: the standard deviation is used to indicate the degree of dispersion of the brain connection distribution across the subject. This finding can indicate the dynamics of FC under fatigue.
  • the difference between the larger subjects in the reorganization, through calculations, the standard deviation of the time-space global efficiency and the time-space local efficiency in the fatigue state is significantly higher than the standard deviation in the awake state. This result is characteristic of the individual
  • the theory is consistent, this feature tends to have the fragility of the accompanying biological matrix over time.
  • the correlation result between the frequency band and the efficiency can be obtained.
  • the nature of the nodes only nodes that show significant fatigue-related differences in each frequency band are selected for correlation testing
  • the embodiment of the present invention introduces a dynamic FC analysis framework and provides an analysis method for studying the spatio-temporal reorganization of dynamic networks during driving fatigue.
  • the small-world attributes existing in the static brain network are extended to the dynamic system.
  • graph theory attributes global efficiency, local efficiency
  • the embodiment of the present invention also provides a driving fatigue-related EEG functional connection dynamic characteristics analysis device.
  • the driving fatigue-related EEG functional connection dynamic characteristics analysis device 1000 includes, but is not limited to: a preprocessing unit 1100, The unit 1200, the space-time topology calculation unit 1300, and the statistical analysis unit 1400.
  • the preprocessing unit 1100 is used to preprocess the EEG data using independent component analysis and wavelet packet transform;
  • the construction unit 1200 is used for constructing the preprocessed EEG data into a time-brain network with dynamic characteristics based on the sliding window method;
  • the spatiotemporal topology calculation unit 1300 is configured to measure the spatiotemporal topology of the time brain network based on the time efficiency analysis framework;
  • the statistical analysis unit 1400 is configured to perform statistical analysis on the spatio-temporal topology of the time-brain network to obtain the correlation between the driving fatigue-related behavior performance and the dynamic characteristics of the time-brain network, and the correlation includes the spatio-temporal global efficiency, The local efficiency of time and space and the centrality of time and space proximity.
  • the embodiment of the present invention also provides a driving fatigue-related EEG functional connection dynamic characteristics analysis device.
  • the driving fatigue-related EEG functional connection dynamic characteristics analysis device 2000 can be any type of smart terminal, such as a mobile phone, a tablet computer, Personal computer, etc.
  • the driving fatigue-related EEG functional connection dynamic characteristic analysis device 2000 includes: one or more control processors 2010 and a memory 2020.
  • one control processor 2010 is taken as an example.
  • control processor 2010 and the memory 2020 may be connected through a bus or in other ways, and the connection through a bus is taken as an example in FIG. 6.
  • the memory 2020 can be used to store non-transitory software programs, non-transitory computer-executable programs and modules, such as driving fatigue-related EEG functional connection dynamic characteristics in the embodiment of the present invention
  • the program instructions/modules corresponding to the analysis method are, for example, the preprocessing unit 1100, the construction unit 1200, the spatiotemporal topology calculation unit 1300, and the statistical analysis unit 1400 shown in FIG. 5.
  • the control processor 2010 runs the non-transitory software programs, instructions and modules stored in the memory 2020 to execute various functional applications and data processing of the driving fatigue-related EEG functional connection dynamic characteristics analysis device 1000, that is, to realize the above method
  • the method for analyzing the dynamic characteristics of the EEG functional connection related to driving fatigue of the embodiment
  • the memory 2020 may include a storage program area and a storage data area, where the storage program area can store an operating system and an application program required by at least one function; the storage data area can store the analysis device 1000 connected with the dynamic characteristics of the EEG function related to driving fatigue Use the created data, etc.
  • the memory 2020 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices.
  • the memory 2020 may optionally include memories remotely provided with respect to the control processor 2010, and these remote memories may be connected to the analysis device 2000 of the dynamic characteristics of the driving fatigue-related EEG function via a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.
  • the one or more modules are stored in the memory 2020, and when executed by the one or more control processors 2010, the method for analyzing the dynamic characteristics of the EEG functional connection related to driving fatigue in the above method embodiment is executed, For example, the method steps S1 to S4 in FIG. 1 described above are executed to realize the functions of the units 1100-1400 in FIG. 5.
  • the embodiment of the present invention also provides a computer-readable storage medium, the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are executed by one or more control processors, for example, as shown in FIG. 6
  • the execution of one control processor 2010 of the above-mentioned one or more control processors 2010 can make the above-mentioned one or more control processors 2010 execute the method of analyzing the dynamic characteristics of driving fatigue-related EEG functional connection in the above-mentioned method embodiment, for example, execute the method in FIG. 1 described above Steps S1 to S4 implement the functions of the units 1100-1400 in FIG. 5.
  • the device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • each implementation manner can be implemented by means of software plus a general hardware platform.
  • All or part of the processes in the methods of the above embodiments can be implemented by computer programs instructing relevant hardware.
  • the programs can be stored in a computer readable storage medium. At this time, it may include the process of the embodiment of the above method.
  • the storage medium may be a magnetic disk, an optical disc, a read-only memory (Read Only Memory, ROM), or a random access memory (Random Access Memory, RAM), etc.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Veterinary Medicine (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Psychiatry (AREA)
  • Psychology (AREA)
  • Developmental Disabilities (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Child & Adolescent Psychology (AREA)
  • Educational Technology (AREA)
  • Hospice & Palliative Care (AREA)
  • Social Psychology (AREA)
  • Signal Processing (AREA)
  • Physiology (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Ophthalmology & Optometry (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

一种驾驶疲劳相关的EEG功能连接动态特性的分析方法,使用独立分量分析和小波包变换对EEG数据进行预处理(S1);基于滑动窗口方法将预处理后的EEG数据构建成动态特性的时间脑网络(S2);基于时间效率分析框架测量时间脑网络的时空拓扑(S3);对时间脑网络的时空拓扑进行统计分析,获得驾驶疲劳相关行为表现与时间脑网络的动态特性之间的相关性(S4),通过将时间特性引入驾驶疲劳的静态网络中,构建具有动态特性的时间脑网络,通过分析统计可以获得驾驶疲劳期间时间脑网络的时空重组规律,具有更准确的分析结果,有利于揭示在精细时间尺度上驾驶疲劳相关的大脑区域间信息传递功能重组的更关键的动力学特性。

Description

一种驾驶疲劳相关的EEG功能连接动态特性的分析方法 技术领域
本发明涉及驾驶疲劳分析技术领域,特别是一种驾驶疲劳相关的EEG功能连接动态特性的分析方法。
背景技术
驾驶疲劳长期以来一直被认为是全球致命事故的主要原因之一,有证据表明,15%-20%的致命交通事故与驾驶疲劳有关,因此近年来研究人员已经在神经人体工程学的新生领域中进行了大量努力以理解驾驶疲劳的神经生物学基础,目的是开发适用的自动检测技术并减少在现实世界中的疲劳相关的交通事故。
目前基于脑电图(EEG)采集实验数据,进行特征提取,再构建与疲劳相关的EEG功能连接(FC)架构是行之有效的方法;过去疲劳研究中的FC为静态连接,即在疲劳状态下以几分钟的时间尺度构建一个代表性脑网络。然而,静态网络研究缺乏关于在精细时间尺度上驾驶疲劳相关的大脑区域间信息传递功能重组的更关键的动力学特性,因此静态FC架构下的实验结果具有一定的局限性。
发明内容
为解决上述问题,本发明的目的在于提供一种驾驶疲劳相关的EEG功能连接动态特性的分析方法,将动态FC分析框架应用到驾驶疲劳研究中,从而获得在精细时间尺度上驾驶疲劳相关的大脑区域间信息传递功能重组的更关键的动力学特性,得到更高的识别准确率。
本发明解决其问题所采用的技术方案是:
一种驾驶疲劳相关的EEG功能连接动态特性的分析方法,包括:
使用独立分量分析和小波包变换对EEG数据进行预处理;
基于滑动窗口方法将预处理后的EEG数据构建成动态特性的时间脑网络;
基于时间效率分析框架测量所述时间脑网络的时空拓扑;
对所述时间脑网络的时空拓扑进行统计分析,获得驾驶疲劳相关行为表现与所述时间脑网络的动态特性之间的相关性,所述相关性包括时空全局效率、时空局部效率和时空邻近中心度。
进一步,所述使用独立分量分析和小波包变换对EEG数据进行预处理包括:
采集眨眼伪影数据,所述眨眼伪影数据包括水平眼电图HEOG数据和垂直眼电图VEOG 数据;
使用独立分量分析查找和删除EEG数据中与所述眨眼伪影数据高度相关的成分;
移除筛选后的EEG数据的基线;
使用小波包变换将所述EEG数据分解为三个标准频段,分别为α频段、β频段和θ频段;
按测试时间将EEG数据划分为清醒状态数据和疲劳状态数据。
进一步,所述基于滑动窗口方法将预处理后的EEG数据构建成动态特性的时间脑网络包括:
将所述预处理后的EEG数据表示成静态网络,所述静态网络为二进制NxN矩阵,其中N表示EEG帽的电极数量;
选择适合的窗口长度和步长,滑动窗口依次遍历整个时间序列,所述时间序列的长度为采集EEG数据的实验时长;
在每个所述静态网络中使用相位滞后指数估算功能连接的PLI值;
采用稀疏度方法,将高于阈值的所述PLI值设置为1,低于阈值的所述PLI值设置为0,从而构成二值化邻接网络,以此作为所述时间脑网络的快照;
将所述静态网络依据时间序列排列构成具有动态特性的时间脑网络。
进一步,所述基于时间效率分析框架测量所述时间脑网络的时空拓扑包括:
在时间尺度上计算成对节点的时间距离,所述时间距离表示被定义为时空路径所经过的时间窗口的最小数量;
计算时空全局效率;
计算时空局部效率;
使用时空邻近中心度评估节点级时间脑网络的时空特征。
一种驾驶疲劳相关的EEG功能连接动态特性的分析装置,包括:
预处理单元,用于使用独立分量分析和小波包变换对EEG数据进行预处理;
构建单元,用于基于滑动窗口方法将预处理后的EEG数据构建成动态特性的时间脑网络;
时空拓扑计算单元,用于基于时间效率分析框架测量所述时间脑网络的时空拓扑;
统计分析单元,用于对所述时间脑网络的时空拓扑进行统计分析,获得驾驶疲劳相关行为表现与所述时间脑网络的动态特性之间的相关性,所述相关性包括时空全局效率、时空局部效率和时空邻近中心度。
进一步,所述预处理单元包括:
眨眼伪影数据采集单元,用于所述眨眼伪影数据包括水平眼电图HEOG数据和垂直眼电图VEOG数据;
筛选单元,用于使用独立分量分析查找和删除EEG数据中与所述眨眼伪影数据高度相关的成分;
基线移除单元,用于移除筛选后的EEG数据的基线;
分解单元,用于使用小波包变换将所述EEG数据分解为三个标准频段,分别为α频段、β频段和θ频段;
数据划分单元,用于按测试时间将EEG数据划分为清醒状态数据和疲劳状态数据。
进一步,所述构建单元包括:
矩阵构建单元,用于将所述预处理后的EEG数据表示成静态网络,所述静态网络为二进制NxN矩阵,其中N表示EEG帽的电极数量;
滑动窗口处理单元,用于选择适合的窗口长度和步长,滑动窗口依次遍历整个时间序列,所述时间序列的长度为采集EEG数据的实验时长;
PLI计算单元,用于在每个所述静态网络中使用相位滞后指数估算功能连接的PLI值;
二值化计算单元,采用稀疏度方法,将高于阈值的所述PLI值设置为1,低于阈值的所述PLI值设置为0,从而构成二值化邻接网络,以此作为所述时间脑网络的快照;
稀疏度计算单元,用于选择合适的稀疏度和间隔,采用稀疏度方法在每个所述静态网络中保留需要的功能连接;
动态特性组建单元,用于将所述静态网络依据时间序列排列构成具有动态特性的时间脑网络。
进一步,所述时空拓扑计算单元包括:
时间距离计算单元,用于在时间尺度上计算成对节点的时间距离,所述时间距离表示被定义为时空路径所经过的时间窗口的最小数量;
时空全局效率计算单元,用于计算时空全局效率;
时空局部效率计算单元,用于计算时空局部效率;
时空特征评估单元,用于使用时空邻近中心度评估节点级时间脑网络的时空特征。
一种驾驶疲劳相关的EEG功能连接动态特性的分析设备,包括至少一个控制处理器和用于与所述至少一个控制处理器通信连接的存储器;所述存储器存储有可被所述至少一个控制处理器执行的指令,所述指令被所述至少一个控制处理器执行,以使所述至少一个控制处理器能够执行如上述任一项所述的驾驶疲劳相关的EEG功能连接动态特性的分析方法。
一种计算机可读存储介质,其特征在于:所述计算机可读存储介质存储有计算机可执 行指令,所述计算机可执行指令用于使计算机执行如上述任一项所述的驾驶疲劳相关的EEG功能连接动态特性的分析方法。
本发明实施例中提供的一个或多个技术方案,至少具有如下有益效果:通过将时间特性引入驾驶疲劳的静态网络中,构建具有动态特性的时间脑网络,通过分析统计可以获得驾驶疲劳期间时间脑网络的时空重组规律,相对于现时驾驶疲劳研究中静态FC连接,本发明的分析方法具有更准确的分析结果,有利于揭示在精细时间尺度上驾驶疲劳相关的大脑区域间信息传递功能重组的更关键的动力学特性。
附图说明
下面结合附图和实施例对本发明作进一步说明。
图1是本发明实施例的整体方法流程图;
图2是本发明实施例的预处理方法的流程图;
图3是本发明实施例的构建动态特性的时间脑网络的流程图;
图4是本发明实施例的测量时间脑网络的时空拓扑的流程图;
图5是本发明实施例的装置中单元架构示意图;
图6是本发明实施例的设备中的连接示意图;
具体实施方式
为了使本发明的目的、技术方案及优点更加清楚明白,以下结合附图及实施例,对本发明进行进一步详细说明。应当理解,此处所描述的具体实施例仅用以解释本发明,并不用于限定本发明。需要说明的是,如果不冲突,本发明实施例中的各个特征可以相互结合,均在本发明的保护范围之内。
需要说明的是,如果不冲突,本发明实施例中的各个特征可以相互结合,均在本发明的保护范围之内。另外,虽然在装置示意图中进行了功能模块划分,在流程图中示出了逻辑顺序,但是在某些情况下,可以以不同于装置中的模块划分,或流程图中的顺序执行所示出或描述的步骤。
参照图1,本发明的一个实施例提供了一种驾驶疲劳相关的EEG功能连接动态特性的分析方法,包括:
S1,使用独立分量分析和小波包变换对EEG数据进行预处理;
S2,基于滑动窗口方法将预处理后的EEG数据构建成动态特性的时间脑网络;
S3,基于时间效率分析框架测量所述时间脑网络的时空拓扑;
S4,对所述时间脑网络的时空拓扑进行统计分析,获得驾驶疲劳相关行为表现与所述时间脑网络的动态特性之间的相关性,所述相关性包括时空全局效率、时空局部效率和时 空邻近中心度。
其中,参照图2,步骤S1包括:
S11,采集眨眼伪影数据,所述眨眼伪影数据包括水平眼电图HEOG数据和垂直眼电图VEOG数据;
S12,使用独立分量分析查找和删除EEG数据中与所述眨眼伪影数据高度相关的成分;
S13,移除筛选后的EEG数据的基线;
S14,使用小波包变换将所述EEG数据分解为三个标准频段,分别为α频段、β频段和θ频段;
S15,按测试时间将EEG数据划分为清醒状态数据和疲劳状态数据。
其中,参照图3,步骤S2包括:
S21,将所述预处理后的EEG数据表示成静态网络,所述静态网络为二进制NxN矩阵,其中N表示EEG帽的电极数量;
S22,选择适合的窗口长度和步长,滑动窗口依次遍历整个时间序列,所述时间序列的长度为采集EEG数据的实验时长;
S23,在每个所述静态网络中使用相位滞后指数估算功能连接的PLI值;
S24,二值化计算单元,采用稀疏度方法,将高于阈值的所述PLI值设置为1,低于阈值的所述PLI值设置为0,从而构成二值化邻接网络,以此作为所述时间脑网络的快照;
S25,将所述静态网络依据时间序列排列构成具有动态特性的时间脑网络。
其中,参照图4,步骤S3包括:
S31,在时间尺度上计算成对节点的时间距离,所述时间距离表示被定义为时空路径所经过的时间窗口的最小数量;
S32,计算时空全局效率;
S33,计算时空局部效率;
S34,使用时空邻近中心度评估节点级时间脑网络的时空特征。
下面根据整体流程对本发明的分析方法进行详细描述:
过去驾驶疲劳研究中的连接(FC)是静态的,即在疲劳状态下以几分钟的时间尺度构建一个代表性脑网络,由于驾驶疲劳的一个关键组成部分是任务时间本身,强调大脑调节疲劳相关性能下降变化的过程的积累特征,因此静态脑网络研究缺乏关于在精细时间尺度上驾驶疲劳相关的大脑区域间信息传递功能重组的更关键的动力学特性。
本发明实施例的实验设置为,受试者进行模拟驾驶90分钟,并在受试者前方设置一辆行驶中的虚拟引导车,虚拟引导车随机间隔产生制动信号,要求受试者通过制动来响应 制动信号以保持安全距离。从虚拟引导车中产生的制动指令到受试者执行制动操作的时间间隔被认为是反应时间(RT),同时还收集车辆的速度变化(SV)作为评估受试者的行为表现的定量指标。
数据采集方式为,受试者佩戴具有改进的国际10-20电极放置系统(HD-72,Cognionics,Inc.,USA)的24通道无线EEG帽以250Hz记录EEG数据,参考电极是右侧和左侧乳突,通过带通滤波器(2到100Hz之间)对EEG信号进行滤波。同时通过放置在眼睛外眼角(水平眼电图,HEOG)和右眼上方和下方的电极(垂直眼电图,VEOG)记录眨眼伪影。
数据预处理:
使用独立分量分析(ICA)来查找和删除与记录的眨眼伪影数据高度相关的成分,在整个实验数据中移除基线,将处理后的数据通过小波包变换(WPT)分解为三个标准频段,分别为θ(3-7Hz)、α(8-13Hz)和β(14-30Hz),在本发明实施例中,小波包变换采用db4和分解级别6的Daubechies小波,来提取脑电信息。
功能连接和时间脑网络构建:
脑网络动态分析框架是用最优滑动窗口表征时间有序的静态网络。在每个静态网络中,使用相位滞后指数(PLI)估算FC,因为它在最小化共源信号和体积传导的影响方面具有优势,FC的PLI值介于0和1之间,数字越大表示连接越强。然后,采用常用的稀疏度方法,选择5%-15%的稀疏度,间隔为1%,将矩阵阈值化为二值化邻接网络,相当于将FC的PLI值与阈值比较,大于阈值的设置为1,小于阈值的设置为0,从而得到二值化邻接网络,以此作为时间脑网络的快照,因此,时间脑网络G={G t}可以由单独的静态网络G t依据时间序列t排列表示,其中t表示正整数,每个静态网络都是一个二进制NxN矩阵,FC的数量与静态网络的数量相同,在本发明实施例的实验中,N的取值为24。
在本发明实施例中窗口长度的选择可以是3-6秒,步长的选择可以是2-4秒,为了平衡信号的动态和连接性估计的质量以及降低计算复杂度,窗口长度选择为4秒,步长选择为4秒,通过对受试者在反应时间和速度变化方面的行为表现进行统计比较来确定最清醒和疲劳状态,最后选择了EEG数据中前5分钟和最后5分钟数据作为分析数据,分别对应最警惕和最疲劳的状态,因此在5分钟内每个窗口的时间步长为T=75。
时间脑网络的时空拓扑:
通过上述构建了动态特性的时间脑网络后,需要在时间尺度上计算成对节点的时间距离,时间距离被定义为时空路径所经过的时间窗口的最小数量,值得注意的是,时间相关路径是空间和时间域的度量,并且时间距离由时域表征。因此,时间距离是正整数,其范围在1和T之间。
为了定量揭示驾驶疲劳期间的大脑动态重组,采用时空效率分析框架来测量时间脑网络的时空拓扑,从概念上讲,时空效率测量动态系统中整个节点之间的交互和信息传递功能,时空全局效率捕获整个网络的动态,信息流的能力在整个生命周期中传播,计算方法如下:
时空全局效率表示为
Figure PCTCN2019109853-appb-000001
Figure PCTCN2019109853-appb-000002
其中G是具有数学结构NxNxT的时空网络,并且
Figure PCTCN2019109853-appb-000003
是时空全局效率在时间t的效率,t的取值为不大于T的正整数。
因此
Figure PCTCN2019109853-appb-000004
Figure PCTCN2019109853-appb-000005
的范围从0到1之间,
Figure PCTCN2019109853-appb-000006
表示所有节点都连接在时空脑网络的快照中,τ i→j(t)表示在时间t从i到j的时间距离。。时空局部效率测量动态网络的整体弹性以在局部范围内随机移除节点:
时空局部效率表示为
Figure PCTCN2019109853-appb-000007
Figure PCTCN2019109853-appb-000008
其中G(i,t)是在时间t包括节点i的所有邻居的子时空网络。
Figure PCTCN2019109853-appb-000009
的范围从0到1之间,它是动态网络的一个索引,用于测量局部规模的信息传播能力。
使用时空邻近中心度评估节点级时间脑网络的时空特征,该中心性测量节点i到达其他节点的能力C c(i,t)表示为:
Figure PCTCN2019109853-appb-000010
时空邻近中心度也表示节点在整个时间网络中的重要性。积分时空接近度中心度表示在整个稀疏度范围内节点i的曲线下面积。
参考网络:
考虑到动态FC中复杂结构的丰富性,需要采用一定的处理方式揭示与参考网络相比动态脑网络的特性和优势。参考网络增加了动态脑网络的随机性,呈现出不同的信息传播 效率。计算这样的参考网络有助于揭示具有不同拓扑的动态FC的神经机制。在本发明实施例中,使用两步随机化方法:随机边(RE)和随机连接(RC)。RE方法在某些约束下随机重新连接动态网络中的所有边缘,这破坏了动态脑网络的拓扑结构。RC方法随机重新分配网络中的所有的连接,这消除了每个边缘的连接数量的分布。应用这两种方法破坏了动态网络的主要结构。将静态网络中的小世界属性的定义扩展到动态特性的时间脑网络,如果满足以下定义,则动态特性的时间脑网络在时空上是小世界的:
Figure PCTCN2019109853-appb-000011
或者
Figure PCTCN2019109853-appb-000012
每个受试者的参考网络的时空效率是在两种心理状态(清醒和疲劳)中具有50次迭代的所生成的时空参考网络的平均值。
统计分析:
为了研究驾驶疲劳对驾驶员控制能力的影响,使用单向重复测量ANOVA来计算整个模拟驾驶任务的反应时间和速度变化,即通过使用单向ANOVA来找出时空全局效率,时空局部效率,清醒和疲劳状态之间的时空邻近中心度的关键差异,为此,进行皮尔逊相关性评估疲劳相关行为表现与动态脑网络特性之间的相关性,这些相关性包括集成的时空全局效率,集成的时空局部效率和集成的时空紧密度中心性,皮尔逊相关性评估以p表示,当p<0.05时被认为是显著相关的。最后通过q=0.05的错误发现率(FDR)进行区域特征的多重比较的校正。
结果分析:
将前后5分钟时间段内反应时间(RT)和速度变化(SV)的值通过坐标的方式分别表示出来,形成时间-反应时间坐标图和时间-速度变化坐标图,可以明显看到最初5分钟和最后5分钟的两种状态之间存在显着差异,从而证实前面定义的最初5分钟为清醒状态,最后5分钟为疲劳状态是合理的。
由此,通过清醒和疲劳状态的时空效率分别定量估计大脑活动的时空拓扑,按照三个标准频带划分建立频带-效率坐标图,来显示显示了整个稀疏范围内的集成时空全局效率和集成时空局部效率,动态特性的时间脑网络在所有三个频带中表现出突出的时空小世界架构:利用标准差表示跨越受试者的脑连接分布的分散程度,该发现可以指示在疲劳状态下动态FC的重组中较大的受试者之间的差异,通过计算可知,疲劳状态下的时空全局效率和时空局部效率的标准偏差明显高于处于清醒状态的标准偏差,这一结果与个体具有特征性特征的理论是一致的,这种特征倾向于具有伴随的生物基质的随时间变化的脆弱性。
时空紧密度中心:
计算24个节点的综合时空邻近中心度,以评估动态FC的节点时空特性,通过计算可知疲劳状态下24个节点的综合时空邻近中心度一般低于清醒状态下的综合时空邻近中心度,同时根据EGG采集的数据对应的脑部区域,可以知道额叶和顶叶中的节点通常显示清醒和疲劳状态之间的巨大差异,即计算结果满足p<0.05。
行为表现与网络属性之间的关系:
通过双变量相关分析研究行为度量ΔRT和ΔSV与时间脑网络的动态特性ΔE之间的关系:
ΔRT=RT fatigue-RT alert
ΔSV=SV fatigue-SV alert
ΔE=E fatigue-E alert
根据分析三个标准频带中的行为度量ΔRT和ΔSV与动态特性ΔE之间的相关性,可以得到频带与效率的相关性结果。对于节点性质,仅选择在每个频带中表现出显着疲劳相关差异的节点进行相关性测试
本发明实施例引入了动态FC分析框架,提供了一种研究驾驶疲劳期间动态网络的时空重组的分析方法,在分析方法的过程中,将静态脑网络中存在的小世界属性扩展到动态系统,结合将时间因素引入FC,并将图论理论属性(全局效率,局部效率)作为特征,获得了比传统分析方法更高的识别准确率,证明了时空架构和时空效率方法在驾驶疲劳检测中的可行性。
本发明实施例还提供了一种驾驶疲劳相关的EEG功能连接动态特性的分析装置,在该驾驶疲劳相关的EEG功能连接动态特性的分析装置1000中,包括但不限于:预处理单元1100、构建单元1200、时空拓扑计算单元1300和统计分析单元1400。
其中,预处理单元1100,用于使用独立分量分析和小波包变换对EEG数据进行预处理;
构建单元1200,用于基于滑动窗口方法将预处理后的EEG数据构建成动态特性的时间脑网络;
时空拓扑计算单元1300,用于基于时间效率分析框架测量所述时间脑网络的时空拓扑;
统计分析单元1400,用于对所述时间脑网络的时空拓扑进行统计分析,获得驾驶疲劳相关行为表现与所述时间脑网络的动态特性之间的相关性,所述相关性包括时空全局效率、时空局部效率和时空邻近中心度。
需要说明的是,由于本实施例中的一种驾驶疲劳相关的EEG功能连接动态特性的分析装置与上述的一种驾驶疲劳相关的EEG功能连接动态特性的分析方法基于相同的发明构思,因此,方法实施例中的相应内容同样适用于本装置实施例,此处不再详述。
本发明实施例还提供了一种驾驶疲劳相关的EEG功能连接动态特性的分析设备,该驾驶疲劳相关的EEG功能连接动态特性的分析设备2000可以是任意类型的智能终端,例如手机、平板电脑、个人计算机等。
具体地,该驾驶疲劳相关的EEG功能连接动态特性的分析设备2000包括:一个或多个控制处理器2010和存储器2020,图6中以一个控制处理器2010为例。
控制处理器2010和存储器2020可以通过总线或者其他方式连接,图6中以通过总线连接为例。
存储器2020作为一种非暂态计算机可读存储介质,可用于存储非暂态软件程序、非暂态性计算机可执行程序以及模块,如本发明实施例中的驾驶疲劳相关的EEG功能连接动态特性的分析方法对应的程序指令/模块,例如,图5中所示的预处理单元1100、构建单元1200、时空拓扑计算单元1300和统计分析单元1400。控制处理器2010通过运行存储在存储器2020中的非暂态软件程序、指令以及模块,从而执行驾驶疲劳相关的EEG功能连接动态特性的分析装置1000的各种功能应用以及数据处理,即实现上述方法实施例的驾驶疲劳相关的EEG功能连接动态特性的分析方法。
存储器2020可以包括存储程序区和存储数据区,其中,存储程序区可存储操作系统、至少一个功能所需要的应用程序;存储数据区可存储根据驾驶疲劳相关的EEG功能连接动态特性的分析装置1000的使用所创建的数据等。此外,存储器2020可以包括高速随机存取存储器,还可以包括非暂态存储器,例如至少一个磁盘存储器件、闪存器件、或其他非暂态固态存储器件。在一些实施方式中,存储器2020可选包括相对于控制处理器2010远程设置的存储器,这些远程存储器可以通过网络连接至该驾驶疲劳相关的EEG功能连接动态特性的分析设备2000。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
所述一个或者多个模块存储在所述存储器2020中,当被所述一个或者多个控制处理器2010执行时,执行上述方法实施例中的驾驶疲劳相关的EEG功能连接动态特性的分析方法,例如,执行以上描述的图1中的方法步骤S1至S4,实现图5中的单元1100-1400的功能。
本发明实施例还提供了一种计算机可读存储介质,所述计算机可读存储介质存储有计算机可执行指令,该计算机可执行指令被一个或多个控制处理器执行,例如,被图6中的一个控制处理器2010执行,可使得上述一个或多个控制处理器2010执行上述方法实施例中的驾驶疲劳相关的EEG功能连接动态特性的分析方法,例如,执行以上描述的图1中的方法步骤S1至S4,实现图5中的单元1100-1400的功能。
以上所描述的装置实施例仅仅是示意性的,其中所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。
通过以上的实施方式的描述,本领域技术人员可以清楚地了解到各实施方式可借助软件加通用硬件平台的方式来实现。本领域技术人员可以理解实现上述实施例方法中的全部或部分流程是可以通过计算机程序来指令相关的硬件来完成,所述的程序可存储于一计算机可读取存储介质中,该程序在执行时,可包括如上述方法的实施例的流程。其中,所述的存储介质可为磁碟、光盘、只读存储记忆体(ReadOnly Memory,ROM)或随机存储记忆体(Random Access Memory,RAM)等。
以上是对本发明的较佳实施进行了具体说明,但本发明并不局限于上述实施方式,熟悉本领域的技术人员在不违背本发明精神的前提下还可作出种种的等同变形或替换,这些等同的变形或替换均包含在本申请权利要求所限定的范围内。

Claims (10)

  1. 一种驾驶疲劳相关的EEG功能连接动态特性的分析方法,其特征在于包括:
    使用独立分量分析和小波包变换对EEG数据进行预处理;
    基于滑动窗口方法将预处理后的EEG数据构建成动态特性的时间脑网络;
    基于时间效率分析框架测量所述时间脑网络的时空拓扑;
    对所述时间脑网络的时空拓扑进行统计分析,获得驾驶疲劳相关行为表现与所述时间脑网络的动态特性之间的相关性,所述相关性包括时空全局效率、时空局部效率和时空邻近中心度。
  2. 根据权利要求1所述的一种驾驶疲劳相关的EEG功能连接动态特性的分析方法,其特征在于,所述使用独立分量分析和小波包变换对EEG数据进行预处理包括:
    采集眨眼伪影数据,所述眨眼伪影数据包括水平眼电图HEOG数据和垂直眼电图VEOG数据;
    使用独立分量分析查找和删除EEG数据中与所述眨眼伪影数据高度相关的成分;
    移除筛选后的EEG数据的基线;
    使用小波包变换将所述EEG数据分解为三个标准频段,分别为α频段、β频段和θ频段;
    按测试时间将EEG数据划分为清醒状态数据和疲劳状态数据。
  3. 根据权利要求1所述的一种驾驶疲劳相关的EEG功能连接动态特性的分析方法,其特征在于,所述基于滑动窗口方法将预处理后的EEG数据构建成动态特性的时间脑网络包括:
    将所述预处理后的EEG数据表示成静态网络,所述静态网络为二进制NxN矩阵,其中N表示EEG帽的电极数量;
    选择适合的窗口长度和步长,滑动窗口依次遍历整个时间序列,所述时间序列的长度为采集EEG数据的实验时长;
    在每个所述静态网络中使用相位滞后指数估算功能连接的PLI值;
    采用稀疏度方法,将高于阈值的所述PLI值设置为1,低于阈值的所述PLI值设置为0,从而构成二值化邻接网络,以此作为所述时间脑网络的快照;
    将所述静态网络依据时间序列排列构成具有动态特性的时间脑网络。
  4. 根据权利要求1所述的一种驾驶疲劳相关的EEG功能连接动态特性的分析方法,其特征在于,所述基于时间效率分析框架测量所述时间脑网络的时空拓扑包括:
    在时间尺度上计算成对节点的时间距离,所述时间距离表示被定义为时空路径所经过 的时间窗口的最小数量;
    计算时空全局效率;
    计算时空局部效率;
    使用时空邻近中心度评估节点级时间脑网络的时空特征。
  5. 一种驾驶疲劳相关的EEG功能连接动态特性的分析装置,其特征在于:包括
    预处理单元,用于使用独立分量分析和小波包变换对EEG数据进行预处理;
    构建单元,用于基于滑动窗口方法将预处理后的EEG数据构建成动态特性的时间脑网络;
    时空拓扑计算单元,用于基于时间效率分析框架测量所述时间脑网络的时空拓扑;
    统计分析单元,用于对所述时间脑网络的时空拓扑进行统计分析,获得驾驶疲劳相关行为表现与所述时间脑网络的动态特性之间的相关性,所述相关性包括时空全局效率、时空局部效率和时空邻近中心度。
  6. 根据权利要求5所述的一种驾驶疲劳相关的EEG功能连接动态特性的分析装置,其特征在于,所述预处理单元包括:
    眨眼伪影数据采集单元,用于所述眨眼伪影数据包括水平眼电图HEOG数据和垂直眼电图VEOG数据;
    筛选单元,用于使用独立分量分析查找和删除EEG数据中与所述眨眼伪影数据高度相关的成分;
    基线移除单元,用于移除筛选后的EEG数据的基线;
    分解单元,用于使用小波包变换将所述EEG数据分解为三个标准频段,分别为α频段、β频段和θ频段;
    数据划分单元,用于按测试时间将EEG数据划分为清醒状态数据和疲劳状态数据。
  7. 根据权利要求5所述的一种驾驶疲劳相关的EEG功能连接动态特性的分析装置,其特征在于,所述构建单元包括:
    矩阵构建单元,用于将所述预处理后的EEG数据表示成静态网络,所述静态网络为二进制NxN矩阵,其中N表示EEG帽的电极数量;
    滑动窗口处理单元,用于选择适合的窗口长度和步长,滑动窗口依次遍历整个时间序列,所述时间序列的长度为采集EEG数据的实验时长;
    PLI计算单元,用于在每个所述静态网络中使用相位滞后指数估算功能连接的PLI值;
    二值化计算单元,采用稀疏度方法,将高于阈值的所述PLI值设置为1,低于阈值的所述PLI值设置为0,从而构成二值化邻接网络,以此作为所述时间脑网络的快照;
    动态特性组建单元,用于将所述静态网络依据时间序列排列构成具有动态特性的时间脑网络。
  8. 根据权利要求5所述的一种驾驶疲劳相关的EEG功能连接动态特性的分析装置,其特征在于,所述时空拓扑计算单元包括:
    时间距离计算单元,用于在时间尺度上计算成对节点的时间距离,所述时间距离表示被定义为时空路径所经过的时间窗口的最小数量;
    时空全局效率计算单元,用于计算时空全局效率;
    时空局部效率计算单元,用于计算时空局部效率;
    时空特征评估单元,用于使用时空邻近中心度评估节点级时间脑网络的时空特征。
  9. 一种驾驶疲劳相关的EEG功能连接动态特性的分析设备,其特征在于:包括至少一个控制处理器和用于与所述至少一个控制处理器通信连接的存储器;所述存储器存储有可被所述至少一个控制处理器执行的指令,所述指令被所述至少一个控制处理器执行,以使所述至少一个控制处理器能够执行如权利要求1-4任一项所述的驾驶疲劳相关的EEG功能连接动态特性的分析方法。
  10. 一种计算机可读存储介质,其特征在于:所述计算机可读存储介质存储有计算机可执行指令,所述计算机可执行指令用于使计算机执行如权利要求1-4任一项所述的驾驶疲劳相关的EEG功能连接动态特性的分析方法。
PCT/CN2019/109853 2019-09-11 2019-10-08 一种驾驶疲劳相关的eeg功能连接动态特性的分析方法 WO2021046949A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US16/975,683 US20210267513A1 (en) 2019-09-11 2019-10-08 Method for analyzing dynamic characteristic of eeg functional connectivity related to driving fatigue
AU2019433218A AU2019433218A1 (en) 2019-09-11 2019-10-08 Method for analyzing dynamic characteristic of EEG Functional connectivity related to driving fatigue

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910859394.9 2019-09-11
CN201910859394.9A CN110584684B (zh) 2019-09-11 2019-09-11 一种驾驶疲劳相关的eeg功能连接动态特性的分析方法

Publications (1)

Publication Number Publication Date
WO2021046949A1 true WO2021046949A1 (zh) 2021-03-18

Family

ID=68859197

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/109853 WO2021046949A1 (zh) 2019-09-11 2019-10-08 一种驾驶疲劳相关的eeg功能连接动态特性的分析方法

Country Status (4)

Country Link
US (1) US20210267513A1 (zh)
CN (1) CN110584684B (zh)
AU (1) AU2019433218A1 (zh)
WO (1) WO2021046949A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114818786A (zh) * 2022-04-06 2022-07-29 五邑大学 通道筛选方法、情绪识别方法、系统及存储介质

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111150392A (zh) * 2020-02-12 2020-05-15 五邑大学 一种有向动态脑功能网络多类情绪识别构建方法及其装置
CN111544017A (zh) * 2020-05-25 2020-08-18 五邑大学 基于gpdc图卷积神经网络的疲劳检测方法、装置及存储介质
CN111882045B (zh) * 2020-08-12 2023-10-17 北京师范大学 基于可微神经结构搜索的大脑时-空网络分解方法及系统
CN112057089A (zh) * 2020-08-31 2020-12-11 五邑大学 情绪识别方法、装置及存储介质
CN113317803B (zh) * 2021-06-04 2022-05-03 重庆邮电大学 一种基于图论和机器学习的神经症特征提取方法
CN113712542A (zh) * 2021-08-27 2021-11-30 中山大学孙逸仙纪念医院 基于深度学习的耳鸣脑电图卷积神经网络模型构建方法
CN114435373B (zh) * 2022-03-16 2023-12-22 一汽解放汽车有限公司 疲劳驾驶检测方法、装置、计算机设备和存储介质
CN114642405A (zh) * 2022-03-22 2022-06-21 南开大学 一种基于动态脑功能网络的帕金森病辅助诊断方法
CN115022185B (zh) * 2022-06-21 2023-09-26 浪潮通信信息系统有限公司 拓扑图显示方法及装置
CN116343487B (zh) * 2023-05-19 2023-08-01 武汉理工大学 兼顾全局效率与局部错位的城市交通路网韧性评估方法
CN116313131B (zh) * 2023-05-24 2023-09-15 山东大学 基于仿造变量的脑网络差异识别系统、设备及存储介质
CN116898401B (zh) * 2023-07-17 2024-02-02 燕山大学 基于联合递归量化的孤独症谱系障碍亚型分类方法及装置

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104050394A (zh) * 2014-07-04 2014-09-17 北京师范大学 群体脑网络的建模方法及其拓扑属性分析方法
US20160090097A1 (en) * 2014-09-29 2016-03-31 The Boeing Company System for fatigue detection using a suite of physiological measurement devices
CN107832656A (zh) * 2017-09-18 2018-03-23 天津大学 一种基于动态功能脑网络的大脑功能状态信息处理方法
CN107909117A (zh) * 2017-09-26 2018-04-13 电子科技大学 一种基于脑功能网络特征对早晚期轻度认知障碍的分类方法及装置
CN108447546A (zh) * 2018-02-02 2018-08-24 常州大学 基于加权网络的大脑功能连接模块划分方法
CN109259764A (zh) * 2018-07-27 2019-01-25 昆明理工大学 一种确定动态脑功能网络阈值方法
CN109770924A (zh) * 2019-01-24 2019-05-21 五邑大学 一种基于广义一致性构建脑功能网络与相关向量机的疲劳分类方法
CN109770923A (zh) * 2019-01-15 2019-05-21 五邑大学 一种基于相位锁定值构建脑功能网络的驾驶疲劳检测方法

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005230030A (ja) * 2004-02-17 2005-09-02 Daikin Ind Ltd 健康管理システム、ヘルメット、処理装置、サーバ装置及び健康管理方法
WO2010089911A1 (ja) * 2009-02-03 2010-08-12 株式会社アクション・リサーチ 振動発生装置及び方法
CN103425983A (zh) * 2013-07-25 2013-12-04 电子科技大学 一种基于网络同步性的脑网络拓扑差异的快速提取方法
CN104473636A (zh) * 2014-12-30 2015-04-01 天津大学 一种基于偏定向相干的疲劳脑网络分析方法
CN107126193B (zh) * 2017-04-20 2020-02-28 杭州电子科技大学 基于滞后阶数自适应选择的多变量因果关系分析方法
CN107095671A (zh) * 2017-05-27 2017-08-29 西南交通大学 驾驶疲劳识别方法及系统
CN107658018B (zh) * 2017-10-12 2019-07-05 太原理工大学 一种基于结构连接和功能连接的融合脑网络构建方法
CN109793528A (zh) * 2019-01-28 2019-05-24 华南理工大学 一种基于动态脑功能网络的情绪分类方法
CN110101384B (zh) * 2019-04-22 2022-01-28 自然资源部第一海洋研究所 用于复杂网络的功能性网络分析系统及分析方法

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104050394A (zh) * 2014-07-04 2014-09-17 北京师范大学 群体脑网络的建模方法及其拓扑属性分析方法
US20160090097A1 (en) * 2014-09-29 2016-03-31 The Boeing Company System for fatigue detection using a suite of physiological measurement devices
CN107832656A (zh) * 2017-09-18 2018-03-23 天津大学 一种基于动态功能脑网络的大脑功能状态信息处理方法
CN107909117A (zh) * 2017-09-26 2018-04-13 电子科技大学 一种基于脑功能网络特征对早晚期轻度认知障碍的分类方法及装置
CN108447546A (zh) * 2018-02-02 2018-08-24 常州大学 基于加权网络的大脑功能连接模块划分方法
CN109259764A (zh) * 2018-07-27 2019-01-25 昆明理工大学 一种确定动态脑功能网络阈值方法
CN109770923A (zh) * 2019-01-15 2019-05-21 五邑大学 一种基于相位锁定值构建脑功能网络的驾驶疲劳检测方法
CN109770924A (zh) * 2019-01-24 2019-05-21 五邑大学 一种基于广义一致性构建脑功能网络与相关向量机的疲劳分类方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114818786A (zh) * 2022-04-06 2022-07-29 五邑大学 通道筛选方法、情绪识别方法、系统及存储介质
CN114818786B (zh) * 2022-04-06 2024-03-01 五邑大学 通道筛选方法、情绪识别方法、系统及存储介质

Also Published As

Publication number Publication date
US20210267513A1 (en) 2021-09-02
AU2019433218A1 (en) 2021-03-25
CN110584684B (zh) 2021-08-10
CN110584684A (zh) 2019-12-20

Similar Documents

Publication Publication Date Title
WO2021046949A1 (zh) 一种驾驶疲劳相关的eeg功能连接动态特性的分析方法
US11538588B2 (en) Atrial fibrillation signal recognition method, apparatus and device
US11232133B2 (en) System for detecting and characterizing seasons
Slanzi et al. Combining eye tracking, pupil dilation and EEG analysis for predicting web users click intention
US11836162B2 (en) Unsupervised method for classifying seasonal patterns
Hosseini et al. Random ensemble learning for EEG classification
WO2021159571A1 (zh) 一种有向动态脑功能网络多类情绪识别构建方法及其装置
Ren et al. Dynamic functional segregation and integration in human brain network during complex tasks
US10699211B2 (en) Supervised method for classifying seasonal patterns
Menshawy et al. An automatic mobile-health based approach for EEG epileptic seizures detection
Rangaprakash et al. Study of phase synchronization in multichannel seizure EEG using nonlinear recurrence measure
Leibig et al. Unsupervised neural spike sorting for high-density microelectrode arrays with convolutive independent component analysis
Rodrigues et al. Space-time recurrences for functional connectivity evaluation and feature extraction in motor imagery brain-computer interfaces
Quaglio et al. Methods for identification of spike patterns in massively parallel spike trains
Dragas et al. Complexity optimization and high-throughput low-latency hardware implementation of a multi-electrode spike-sorting algorithm
CN107095684A (zh) 一种基于脑电的儿童自闭症风险评估系统
Labate et al. EEG complexity modifications and altered compressibility in mild cognitive impairment and Alzheimer’s disease
Karpov et al. Extreme value theory inspires explainable machine learning approach for seizure detection
Hamatta et al. Genetic Algorithm‐Based Human Mental Stress Detection and Alerting in Internet of Things
Klepl et al. Cross-frequency multilayer network analysis with bispectrum-based functional connectivity: A study of Alzheimer’s disease
Callegari et al. EpiCare—A home care platform based on mobile cloud computing to assist epilepsy diagnosis
Kathole et al. Integrating MLOps and EEG Techniques for Enhanced Crime Detection and Prevention
EP3745947A1 (en) Method for detecting a conversion from mild cognitive impairment to alzheimer disease
Nguyen et al. HOKF: High order kalman filter for epilepsy forecasting modeling
Antonakakis et al. Mining cross-frequency coupling microstates from resting state MEG: an application to mild traumatic brain injury

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2019433218

Country of ref document: AU

Date of ref document: 20191008

Kind code of ref document: A

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19944803

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19944803

Country of ref document: EP

Kind code of ref document: A1