CN110192860A - A kind of the Brian Imaging intelligent test analyzing method and system of network-oriented information cognition - Google Patents
A kind of the Brian Imaging intelligent test analyzing method and system of network-oriented information cognition Download PDFInfo
- Publication number
- CN110192860A CN110192860A CN201910372890.1A CN201910372890A CN110192860A CN 110192860 A CN110192860 A CN 110192860A CN 201910372890 A CN201910372890 A CN 201910372890A CN 110192860 A CN110192860 A CN 110192860A
- Authority
- CN
- China
- Prior art keywords
- test
- data
- network
- network information
- imaging
- 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.)
- Granted
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves
- A61B5/055—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
-
- 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
- A61B5/7242—Details of waveform analysis using integration
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Radiology & Medical Imaging (AREA)
- High Energy & Nuclear Physics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention belongs to information technology field, the Brian Imaging intelligent test analyzing method and system of specially a kind of network-oriented information cognition.The method of the present invention includes network information recognition tests pretreatment, the test of network information Cognitive task, the cognition of Brian Imaging data calculates, network information cognition result exports 4 steps；Detecting and analysing system of the present invention includes four software modules and the network information multimedia presentation module, cognitive reaction signal acquisition module, 4 functional magnetic resonance imaging Brian Imaging module, intelligent analysis module hardware modules corresponding to 4 steps.The invention has the advantages that testing signal using virtual reality technology planned network information；Based on neural network model to brain function network modelling；Brain tranquillization state, DTI, f-MRI test data are subjected to convergence analysis；Intelligence computation is realized by artificial intelligence technology.More accurately Measurement results can be obtained using the present invention.
Description
Technical field
The invention belongs to information technology fields, and in particular to a kind of Brian Imaging intelligence test of network-oriented information cognition point
Analyse method and system.
Background technique
With becoming increasingly popular for Internet application, the market demand is held by the cognitive analysis to the network information, is understood
The psychology and behavioural characteristic of consumer has become enterprise and works out marketing strategy, improves product design and production method, promotion clothes
The new way for level of being engaged in, it is competing for promoting the intelligence s ervice of the state keys industry developments such as intelligence manufacture and promotion modern service industry
The ability of striving is of great significance.At the same time, the cognitive analysis of the network information, which has become, knows Social Psychology clearly, studies and judges public sentiment trend
A key technology, play significant role in sudden incidents report, national security and governance.
Network information cognition includes the heart of a series of complex such as the feeling to above- mentioned information, consciousness, memory, the imagination, thinking
Manage active procedure, mainly have two major classes about the method for testing and analyzing of cognition at present: one kind is the subjectivity using Experiment of Psychology
Readme evaluating method carries out fuzzy evaluation and test analysis by the subjective experience readme recognized to the network information；It is another kind of to cognition
The physiological signal of reaction and external manifestation, such as brain electricity, electrocardio, breathing, expression, posture, movement, are acquired calculating, are based on
Above data carries out evaluation and test analysis.In the above-mentioned methods, subjective readme evaluating method is because resonable to subjective experience by subject
Solution, the differentia influence in statement are there is compared with big limitation, the calculating evaluating method based on physiological signal Yu external manifestation data
Since there is nonuniquenesses and insufficiency between cognitive reaction and above-mentioned physiological signal, external manifestation, it is difficult to which it is accurate to obtain
Evaluation result.
Functional magnetic resonance imaging Brian Imaging (f-MRI) is a kind of emerging neuroimaging technology, can be with the lossless detection external world
The cerebral nerve activity that signal stimulus is caused, is increasingly being applied in cognitive analysis in recent years.However, the network information
Appearance form it is complicated, be related to the combination of the multimedia messages such as text, audio, video and its complexity, to above- mentioned information
The method for testing and analyzing system of cognition problem to be resolved.
Summary of the invention
The purpose of the present invention is in view of the deficiencies of the prior art, propose a kind of Brian Imaging intelligence of network-oriented information cognition
Method for testing and analyzing and system, the calculating by artificial intelligence technology to brain function network modelling and its characteristic parameter are above-mentioned
The test analysis of information cognition provides more accurately method and system scheme.
The Brian Imaging intelligent test analyzing method of network-oriented information cognition proposed by the present invention, including following four are specific
Step:
Step 1: network information recognition tests pretreatment；It is collected including network information recognition tests material, subject personnel survey
Trial signal design and 3 processes of Brian Imaging tester parameter setting, in which:
Network information recognition tests material is collected: being collected all kinds of texts, audio, video material information from internet, is used for
Test the design of signal；
Subject personnel test Design of Signal: first according to the network information recognition tests material being collected into, using virtual existing
Real technology carries out compilation and design to the presentation mode of the static state of material, dynamic and background information, then passes through wearable eyeshade
Design scheme is presented, is observed based on portable event related potential ERPs information, selection is most able to satisfy test assignment and wants
The scheme asked finally provides the specific presentation mode of test signal of above scheme in the case where stimulation software is presented in E-prime, comprising:
Establish instruction, the time is presented in setting, setting response mode, the imaginary idea activity for prompting subject personnel, prompt subject personnel
Eye movement and lip reading, prompt subject personnel to the operation of keyboard, provide conclusion；
Brian Imaging tester parameter setting: being configured the parameter of functional magnetic resonance Brian Imaging instrument, including imaging ginseng
Number selection, region of interest selection, debugging calibration；
Step 2: the test of network information Cognitive task；It is executed including subject personnel's test assignment, subject personnel recognize data
Acquisition and subject personnel recognize data processing totally 3 processes, in which:
Subject personnel's test assignment executes: personnel are in test cabinet for subject, drawn according to the presentation of test signal and its prompt
It leads, completes test assignment and corresponding operation movement；
Subject personnel recognize data acquisition: during the test, acquire subject brain tranquillization state imaging, diffusion tensor at
Picture DTI and Magnetic resonance imaging f-MRI is total to three classes test data, and is transmitted to together with the synchronizing information of test signal
Recognize data acquisition computer；
Subject personnel recognize data processing: being corrected to collected test data and marking is handled；
Step 3: the network information recognizes intelligence computation；Including brain function network modelling, the building of cognitive knowledge library, Brian Imaging number
According to intelligence computation totally 3 processes, in which:
Brain function network modelling: the nerve fibre distribution characteristics based on diffusion tensor imaging establishes brain function network model
And the basic connection relationship between each functional areas, the initial value of above-mentioned connection relationship is calculated by brain tranquillization state imaging data；
Brain function network model is expressed using following neutral net mathematical model:
yiFor the neuron models output of i-th of brain area, θiFor the threshold value of above-mentioned output；βijFor j-th of brain area and i-th
Nerve fibre coefficient of connection between brain area, is calculated by DTI data；θ0For the benchmark correction value of threshold value, wijFor j-th brain area with
Brain function network weight coefficient between i-th of brain area, initial value and θ0Provided by the calculating of brain tranquillization state imaging data；xj
For the activation level of j-th of brain area, calculated by f-MRI data；
Cognitive knowledge library building: in above-mentioned brain function network models, using deep neural network machine learning skill
Art is trained the relationship between Magnetic resonance imaging data and the network information cognition result for calibration, and mentions
Characteristic parameter of the brain function network parameter for taking its stable as network information cognitive analysis, construct cognitive knowledge library, will more than
Characteristic parameter is stored in knowledge base；
Brian Imaging data intelligence calculate: to the Magnetic resonance imaging data obtained in different test assignments with deposit
The brain function network characterization parameter in cognitive knowledge library is stored up, as neuron models export yiAnd its it is time series parameters, average
Path length li, component efficiency Elocal, global efficiency Eglobal, matching primitives are carried out, the network information is obtained and recognizes result；
Step 4: the network information recognizes result output；It exports in different test assignments to the calculating knot of network information cognition
Fruit, and this test result is used for by trigger the update in cognitive knowledge library.
The Brian Imaging intelligence test analysis system of network-oriented information cognition proposed by the present invention, including following four software
Module:
Network information recognition tests pretreatment, including network information recognition tests material is collected, subject personnel test signal
Design and 3 submodules of Brian Imaging tester parameter setting, this 3 submodules execute the function of 3 processes in 1 step 1 respectively；
The network information Cognitive task test, including subject personnel's test assignment execute, subject personnel recognize data acquisition and
Subject personnel recognize 3 submodules of data processing, this 3 submodules execute the function of 3 processes in step 2 respectively；
The network information recognizes intelligence computation, including brain function network modelling, the building of cognitive knowledge library, Brian Imaging data intelligence
3 submodules are calculated, this 3 submodules execute the function of 3 processes in step 3 respectively；
The network information recognizes result output, the function of this module step 4.
The Brian Imaging intelligence test analysis system of network-oriented information cognition proposed by the present invention, further includes that following four is hard
Part module:
Module is presented in network information multimedia: for network information recognition tests signal to be presented to subject personnel, by network
Information material database, stimulus signal instrument and network information multimedia presentation devices are total to three parts composition；
Cognitive reaction signal acquisition module: it for acquiring the cognitive reaction signal of subject personnel, is adopted by cognitive reaction signal
Acquisition means and synchronization signal measure and control device are total to two parts composition；
Functional magnetic resonance imaging Brian Imaging module: for acquiring the dynamic Brian Imaging data of subject personnel, it is by test number
Three parts composition is total to according to library, experiment Measurement &control computer and functional magnetic resonance imaging Brian Imaging instrument；
Intelligent analysis module: for configuring the software module for realizing 4 steps in claim 1, and being above-mentioned module
Operation provides soft and hardware working environment；
The detecting and analysing system being made of above four modules, operating mode include following four process:
Process 1: start to test；
Tranquillization state and DTI test instruction are issued by experiment Measurement &control computer, sent out after the processing of synchronization signal measure and control device
Signal is controlled out, and network information multimedia presentation devices is made to be in the quiescent condition presented without any signal；
Process 2: tranquillization state data acquisition；
Startup function nuclear magnetic resonance Brian Imaging instrument to subject carry out brain scanning, acquire corresponding brain tranquillization state data and
DTI data, and send above-mentioned test data to experiment Measurement &control computer, above-mentioned data are read by special-purpose software in the computer
After store in test database；
Process 3: synchronous data collection；
After complete process 2, cognitive reaction and f-MRI test instruction are issued by experiment Measurement &control computer, by synchronization
Control signal is issued after the processing of signal measure and control device, starts network information multimedia presentation devices, and according to scheduled signal
Step sequence is presented and plays testing stimulus signal, meanwhile, pass through cognitive reaction signal pickup assembly, functional magnetic resonance imaging Brian Imaging instrument
Acquisition is synchronized to subjective judgement information of the subject under corresponding stimulus signal and its f-MRI brain scanning data, synchronizes and adopts
Above-mentioned test data is sent to experiment Measurement &control computer after the completion of collection, after reading above-mentioned data by special-purpose software in the computer
It stores in test database；
Process 4: test terminates；
After above-mentioned whole test process are completed, instruction is completed by the sending of experiment Measurement &control computer, makes each test
Equipment is restored to the original state before test.
In the present invention, above four software modules are configured in the intelligent analysis module of hardware system, can be with hardware system
Network information multimedia present module, functional magnetic resonance imaging Brian Imaging module carry out data communication.
The present invention provides more accurately scheme for the test analysis that the network information recognizes, in which: uses virtual existing
The real Technology design network information tests signal, and presented by wearable eyeshade, portable event related potential ERPs,
Observation and optimization design in advance improve the designing quality and validity of test signal；Based on neutral net mathematical model come pair
Brain function network modelling, compared with traditional statistical modeling method, the complicated dynamic that can preferably embody brain function network is special
Sign；Brain tranquillization state, DTI, f-MRI test data are subjected to convergence analysis, more accurately test result can be obtained；Pass through people
Work intellectual technology is calculated, is identified, the intelligence of Data Management Analysis is realized.
Detailed description of the invention
System global structure figure Fig. 1 of the invention.
Test analysis frame diagram Fig. 2 of the invention.
System operating mode figure Fig. 3 of the invention.
Specific embodiment
With reference to the accompanying drawings, various implementations of the invention are described in further detail.
1, Fig. 1 shows system global structure figure of the invention.
Module, cognitive reaction is presented by network information multimedia in detecting and analysing system of the present invention, overall structure
Totally 4 modules form for signal acquisition module, functional magnetic resonance imaging Brian Imaging module, intelligent analysis module.Wherein, intellectual analysis
It is configured with software module corresponding with four steps of method for testing and analyzing of the present invention are realized in module, and is above-mentioned module
Operation provide soft and hardware working environment.Synchronization signal measure and control device and the network information in cognitive reaction signal acquisition module
Multimedia presentation devices, cognitive reaction signal pickup assembly, Magnetic resonance imaging instrument are attached communication, realize signal
Synchronous and observing and controlling.Module is presented with network information multimedia in intelligent analysis module, cognitive reaction signal acquisition module is attached
Communication, realizes the software function of four steps of method for testing and analyzing of the present invention.
2, Fig. 2 shows test analysis frame diagrams of the invention.
Method for testing and analyzing of the present invention, frame are recognized by network information recognition tests pretreatment, the network information
Task test, network information cognition intelligence computation, network information cognition result export four step compositions, are implemented as follows:
Step 1: collecting all kinds of texts, audio, video material information from internet, such as: the finance from People's Net's public sentiment
Public sentiment observes column, education public sentiment observation column, medical public sentiment observation column, environmentally friendly public sentiment observation column and collects information, from new
The network public-opinion (weekly) of China's net collects information, collects information from the network forum of all kinds of social accidents, using virtual existing
Real technology carries out compilation and design to the presentation mode of the static state of material, dynamic and background information, presents in wearable eyeshade,
The lower event-related EEG reaction induced is stimulated to be observed above- mentioned information by portable brain Evoked ptential ERPs, according to it
Significant difference selection is most able to satisfy the scheme of test assignment requirement, provides above scheme in the case where stimulation software is presented in E-prime
Test signal specific presentation mode；Then, the parameter of functional MRI Brian Imaging instrument is configured.
Step 2: first test assignment and test points for attention are informed to subject personnel before test, with subject personnel's signature
Informed consent form；Later, subject personnel enter test cabinet, lie on testboard, believe according to test assignment design scheme in network
It ceases and test signal is presented on multimedia presentation devices, subject personnel complete corresponding according to the prompt information occurred in above-mentioned signal
Operational motion, such as: opening eyes, close one's eyes, reading lip reading, illusion idea, keyboard operation selection silently；Pass through cognitive reaction signal pickup assembly
Subject's keyboard operation during the test and all kinds of cognitive reaction physiological signals are acquired, and by synchronization signal observing and controlling
Above data is sent in the test database of functional magnetic resonance imaging Brian Imaging module by device together with synchronization signal；It is logical
Functional magnetic resonance imaging Brian Imaging instrument is crossed to the imaging of brain tranquillization state, diffusion tensor imaging (DTI) and the functional magnetic resonance of subject
Resonance image-forming (f-MRI) is total to three classes test data and is acquired, and above data is stored in functional magnetic resonance imaging Brian Imaging mould
In the test database of block；To collected Brian Imaging data, it is corrected and marking is handled, specifically first select piece image
As benchmark image, then by translating or rotating angle for the position of other images and piece image in time series
Match, then by interpolation algorithm to all image resamplings, and then completes the change in location on head and the school of angle change
Just, however the function image of every subject is matched in its structural images, forms MNI standardization after specification handles
Data.
Step 3: on the basis of human brain anatomical template AAL (Automated Anantomical Labeling), by brain
It is divided into 90 brain areas, each brain area is defined as a node of brain function network, is built using following neutral net mathematical model
Mould:
The DTI data of diffusion tensor imaging are handled, its nerve fibre bundle tracking image is obtained, calculate Different brain region
Between nerve fibre bundle connect quantity, then according to following formula calculate β ij:
βij=FBij/FBmax
Wherein, FBijNerve fibre bundle between j brain area and i brain area connects quantity, and FBmax is between each brain area
Nerve fibre bundle connect maximum quantity, obtain calculated result such as: β21,22=0.1021；
The f-MRI data being imaged by brain tranquillization state, using Pearson correlation coefficient and time series residual computations brain function
It can network weight coefficient wijInitial value and neuron models input threshold value benchmark correction value, such as power between the 21st, 22 brain areas
Coefficient initial value is w21,22The benchmark correction value of the neuron models input threshold value of=0.3134, the 21st brain area is θ21=
0.0742；
Based on the above brain function network model, using deep neural network machine learning techniques to the functionality for calibration
Relationship between nmr imaging data and network information cognition result is trained, the specific implementation process is as follows:
The above deep neural network machine learning is by a series of limited Boltzmann machine (Restricted Boltzmann
Machine, RBM) composition is stacked, learnt in such a way that unsupervised greediness is successively trained, RBM is an energy model,
Its energy function is defined as follows:
In formula, v is the state vector of visible layer unit, and h is the state vector of implicit layer unit, viIt is the i-th of visible layer
The state of a node, hjIt is the state of j-th of node of hidden layer, wijIt is i-th of visible node layer and j-th of hidden layer node
Connection weight, it is seen that the joint probability distribution of node layer and hidden layer node is as follows:
In above formula, u is the state of all visible layers, and g is the state of all possible hidden layer, and v is given visible layer
State, h are the states of given hidden layer, and molecule is energy possessed by current state (v, h), and denominator is all possible states
It to the energy of (u, g), is calculated by above-mentioned probabilistic model when given visible layer inputs, the probability that each node of hidden layer is activated
And output state, if data data are the input state of hidden layer and visible layer, the output shape of corresponding hidden layer and visible layer
State is denoted as recon, then finally obtains weight coefficient more new formula using the multiple Gibbs method of sampling are as follows:
Δwij=ε (< vihj>data-<vihj>recon)
After the training of deep neural network machine learning techniques, obtain corresponding with heterogeneous networks information cognition result
Brain function Network characteristic parameters, as neuron models export yiAnd its time series parameters, average path length li, part effect
Rate Elocal, global efficiency Eglobal, above-mentioned parameter is stored in the cognitive knowledge library of building；
Using support vector machines multidimensional sorting algorithm, to the Magnetic resonance imaging obtained in different test assignments
Data carry out matching primitives with the brain function network characterization parameter being stored in cognitive knowledge library, obtain network information cognition knot
Fruit, the specific implementation process is as follows:
Let R be the set as composed by the N number of network information cognition result stored in knowledge base, FiFor one group and wherein the
The corresponding brain function network characterization parameter of i cognition result, M are Magnetic resonance imaging f-MRI data, pass through support
Brain function network parameter and F of the vector machine algorithm to Mi(i=1,2,3 ..., N) carries out matching primitives respectively, and distance is minimum
Result be recognition result, if Cognitive task is that " front " recognized to the network information, " neutrality ", " negative " attitude are made and being sentenced
It is disconnected, and support vector machines multidimensional sorting algorithm is calculated is respectively 0.2561 at a distance from above-mentioned three classes result, 0.7314,
0.5212, then recognition result is " neutrality ", and recognition result can also be expressed is the percentage at a distance from above-mentioned three classes attitude, i.e.,
" front " 16.97%, " neutrality " 48.48%, " negative " 34.55%.
Step 4: by network information cognition result output, and this test result being used for by cognitive knowledge library by trigger
Update.
3, Fig. 2 shows system operating mode figures of the invention.
Process 11: when test starts, tranquillization state and DTI test instruction are issued by experiment Measurement &control computer first；
Process 12: above-metioned instruction sets network information multimedia presentation devices to the tranquillization shape presented without any signal
State；
Process 13: startup function nuclear magnetic resonance Brian Imaging instrument carries out brain scanning to subject, acquires corresponding brain tranquillization
State data and DTI data；
Process 14: the collected data transmission of process 3 is given to experiment Measurement &control computer；
Process 15: storing data into test database and judges that tranquillization state tests whether to finish, if do not tested,
Return to process 12；Enter process 16 if having tested；
Process 16: experiment Measurement &control computer issues cognitive reaction and f-MRI test instruction；
Process 17: synchronization signal measure and control device issues control signal enabling network information multimedia presentation devices；
Process 18: network information multimedia presentation devices play testing stimulus signal according to the presentation mode of setting；
Process 19: it is tested after the stimulus signal that viewing process 8 is showed, is provided by cognitive reaction signal pickup assembly
Subjective response under the stimulus signal；
Process 20: the stimulus signal that functional magnetic resonance imaging Brian Imaging instrument scanning collection subject is showed in viewing process 8
When fMRI data；
Process 21: above-mentioned test data is sent to experiment Measurement &control computer again；
Process 22: storing data into test database and judges that f-MRI tests whether to finish, if do not tested,
Return to process 18；Enter process 23 if having tested；
Process 23: experiment Measurement &control computer sending is completed instruction, initial before so that each test equipment is restored to test
State.
Above embodiment is preferred case of the present invention in common situation, the protection model being not intended to limit the invention
It encloses.
Claims (4)
1. a kind of Brian Imaging intelligent test analyzing method of network-oriented information cognition, which is characterized in that specific step is as follows:
Step 1: network information recognition tests pretreatment；It is collected including network information recognition tests material, subject personnel test letter
Number design and 3 processes of Brian Imaging tester parameter setting, in which:
Network information recognition tests material is collected: all kinds of texts, audio, video material information is collected from internet, for testing
The design of signal；
Subject personnel test Design of Signal: first according to the network information recognition tests material being collected into, using virtual reality skill
Art carries out compilation and design to the presentation mode of the static state of material, dynamic and background information, then by wearable eyeshade to setting
Meter scheme is presented, and is observed based on portable event related potential ERPs information, and selection is most able to satisfy test assignment requirement
Scheme finally provides the specific presentation mode of test signal of above scheme, comprising: establish in the case where stimulation software is presented in E-prime
Time, setting response mode, the imaginary idea activity for prompting subject personnel, the eye for prompting subject personnel is presented in instruction, setting
Dynamic and lip reading, prompt subject personnel to the operation of keyboard, provide conclusion；
Brian Imaging tester parameter setting: being configured the parameter of functional magnetic resonance Brian Imaging instrument, including imaging parameters choosing
It selects, region of interest selection, debugging calibration；
Step 2: the test of network information Cognitive task；It is executed including subject personnel's test assignment, subject personnel recognize data acquisition
Data processing totally 3 processes are recognized with subject personnel, in which:
Subject personnel's test assignment executes: personnel are in test cabinet for subject, complete according to the presentation of test signal and its prompting and guiding
It is acted at test assignment and corresponding operation；
Subject personnel recognize data acquisition: during the test, acquiring the imaging of brain tranquillization state, the diffusion tensor imaging of subject
DTI and Magnetic resonance imaging f-MRI is total to three classes test data, and is transmitted to and recognizes together with the synchronizing information of test signal
Primary data collecting computer；
Subject personnel recognize data processing: being corrected to collected test data and marking is handled；
Step 3: the network information recognizes intelligence computation；Including brain function network modelling, the building of cognitive knowledge library, Brian Imaging data intelligence
Totally 3 processes can be calculated, in which:
Brain function network modelling: the nerve fibre distribution characteristics based on diffusion tensor imaging, establish brain function network model and
Basic connection relationship between each functional areas calculates the initial value of above-mentioned connection relationship by brain tranquillization state imaging data；
Brain function network model is expressed using following neutral net mathematical model:
y i It isiThe neuron models of a brain area export,θ i For the threshold value of above-mentioned output；β ij It isjA brain area and theiA brain area
Between nerve fibre coefficient of connection, calculated by DTI data；θ 0 For the benchmark correction value of threshold value,ω ij It isjA brain area and thei
Brain function network weight coefficient between a brain area, initial value andθ 0 Provided by the calculating of brain tranquillization state imaging data；x j It isjThe activation level of a brain area is calculated by f-MRI data；
Cognitive knowledge library building: in above-mentioned brain function network models, using deep neural network machine learning techniques pair
It is trained for the relationship between Magnetic resonance imaging data and the network information cognition result of calibration, and extracts it
Characteristic parameter of the stable brain function network parameter as network information cognitive analysis constructs cognitive knowledge library, by features above
Parameter is stored in knowledge base；
Brian Imaging data intelligence calculate: to the Magnetic resonance imaging data obtained in different test assignments be stored in
Brain function network characterization parameter in cognitive knowledge library, including neuron models outputy i And its time series parameters, average road
Electrical path lengthl i , component efficiencyE local , global efficiencyE global , matching primitives are carried out, the network information is obtained and recognizes result；
Step 4: the network information recognizes result output；The calculated result recognized in different test assignments to the network information is exported,
And this test result to be used for the update in cognitive knowledge library by trigger.
2. the Brian Imaging intelligence test analysis system of the network-oriented information cognition based on method described in claim 1, special
Sign is, including following four software module:
Network information recognition tests preprocessing module, including network information recognition tests material is collected, subject personnel test signal
Design and 3 submodules of Brian Imaging tester parameter setting, this 3 submodule difference perform claims require 3 mistakes in 1 step 1
The function of journey；
Network information Cognitive task test module, including subject personnel's test assignment execute, subject personnel recognize data acquisition and
Subject personnel recognize 3 submodules of data processing, this 3 submodule difference perform claims require the function of 3 processes in 1 step 2
Energy；
The network information recognizes intelligence computation module, including brain function network modelling, the building of cognitive knowledge library, Brian Imaging data intelligence
3 submodules are calculated, this 3 submodule difference perform claims require the function of 3 processes in 1 step 3；
The network information recognizes result output module, this module perform claim requires the function of 1 step 4.
3. the Brian Imaging intelligence test analysis system of the network-oriented information cognition of the method according to claim 11, special
Sign is, further includes following four hardware module:
Module is presented in network information multimedia: for network information recognition tests signal to be presented to subject personnel, by the network information
Material database, stimulus signal instrument and network information multimedia presentation devices are total to three parts composition；
Cognitive reaction signal acquisition module: it for acquiring the cognitive reaction signal of subject personnel, is filled by cognitive reaction signal acquisition
It sets and is total to two parts composition with synchronization signal measure and control device；
Functional magnetic resonance imaging Brian Imaging module: for acquiring the dynamic Brian Imaging data of subject personnel, it by test database,
Experiment Measurement &control computer and functional magnetic resonance imaging Brian Imaging instrument are total to three parts composition；
Intelligent analysis module: for configuring the software module for realizing 4 steps in claim 1, and being the operation of above-mentioned module
Soft and hardware working environment is provided；
The detecting and analysing system being made of above four modules, operating mode include following four process:
Process 1: start to test；
Tranquillization state and DTI test instruction are issued by experiment Measurement &control computer, issue control after the processing of synchronization signal measure and control device
Signal processed makes network information multimedia presentation devices be in the quiescent condition presented without any signal；
Process 2: tranquillization state data acquisition；
Startup function nuclear magnetic resonance Brian Imaging instrument carries out brain scanning to subject, acquires corresponding brain tranquillization state data and DTI
Data, and send above-mentioned test data to experiment Measurement &control computer, after reading above-mentioned data by special-purpose software in the computer
It stores in test database；
Process 3: synchronous data collection；
After complete process 2, cognitive reaction and f-MRI test instruction are issued by experiment Measurement &control computer, by synchronization signal
Control signal is issued after measure and control device processing, starts network information multimedia presentation devices, and is presented according to scheduled signal
It walks sequence and plays testing stimulus signal, meanwhile, by cognitive reaction signal pickup assembly, functional magnetic resonance imaging Brian Imaging instrument to quilt
Subjective judgement information and its f-MRI brain scanning data of the examination person under corresponding stimulus signal synchronize acquisition, and synchronous acquisition is complete
Above-mentioned test data is sent to experiment Measurement &control computer after, is stored after reading above-mentioned data by special-purpose software in the computer
Into test database；
Process 4: test terminates；
After above-mentioned whole test process are completed, instruction is completed by the sending of experiment Measurement &control computer, makes each test equipment
Original state before being restored to test.
4. the Brian Imaging intelligence test analysis system of network-oriented information cognition according to claim 3, which is characterized in that
Four software modules are configured in the intelligent analysis module of hardware system, mould can be presented with the network information multimedia of hardware system
Block, functional magnetic resonance imaging Brian Imaging module carry out data communication.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910372890.1A CN110192860B (en) | 2019-05-06 | 2019-05-06 | Brain imaging intelligent test analysis method and system for network information cognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910372890.1A CN110192860B (en) | 2019-05-06 | 2019-05-06 | Brain imaging intelligent test analysis method and system for network information cognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110192860A true CN110192860A (en) | 2019-09-03 |
| CN110192860B CN110192860B (en) | 2022-10-11 |
Family
ID=67752367
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910372890.1A Active CN110192860B (en) | 2019-05-06 | 2019-05-06 | Brain imaging intelligent test analysis method and system for network information cognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110192860B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111539509A (en) * | 2020-04-16 | 2020-08-14 | 山东第一医科大学(山东省医学科学院) | Virtual digital brain construction method and system and intelligent robot control system |
| CN111861115A (en) * | 2020-06-15 | 2020-10-30 | 复旦大学 | Neural experiment test calculation method for consumer cognition of internet financial products |
| CN113177867A (en) * | 2021-04-22 | 2021-07-27 | 行云新能科技（深圳）有限公司 | Teaching method and device of chassis line control system and readable storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140222738A1 (en) * | 2011-06-09 | 2014-08-07 | Karen E. Joyce | Agent-Based Brain Model and Related Methods |
| CN104361203A (en) * | 2014-10-17 | 2015-02-18 | 东南大学 | Social emotion competency evaluation system based on brain network time-space dynamics analysis |
| CN106056602A (en) * | 2016-05-27 | 2016-10-26 | 中国人民解放军信息工程大学 | CNN (convolutional neural network)-based fMRI (functional magnetic resonance imaging) visual function data object extraction method |
| CN107045593A (en) * | 2017-04-17 | 2017-08-15 | 浙江大学 | Brain function experimental duties stimulating system with Real-time Feedback and task more New function |
| CN108056774A (en) * | 2017-12-29 | 2018-05-22 | 中国人民解放军战略支援部队信息工程大学 | Experimental paradigm mood analysis implementation method and its device based on visual transmission material |
| CN109034360A (en) * | 2018-07-13 | 2018-12-18 | 北京工业大学 | A kind of ant colony method constructing brain effective connectivity network from fMRI and DTI data |
| CN109222972A (en) * | 2018-09-11 | 2019-01-18 | 华南理工大学 | A kind of full brain data classification method of fMRI based on deep learning |
-
2019
- 2019-05-06 CN CN201910372890.1A patent/CN110192860B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140222738A1 (en) * | 2011-06-09 | 2014-08-07 | Karen E. Joyce | Agent-Based Brain Model and Related Methods |
| CN104361203A (en) * | 2014-10-17 | 2015-02-18 | 东南大学 | Social emotion competency evaluation system based on brain network time-space dynamics analysis |
| CN106056602A (en) * | 2016-05-27 | 2016-10-26 | 中国人民解放军信息工程大学 | CNN (convolutional neural network)-based fMRI (functional magnetic resonance imaging) visual function data object extraction method |
| CN107045593A (en) * | 2017-04-17 | 2017-08-15 | 浙江大学 | Brain function experimental duties stimulating system with Real-time Feedback and task more New function |
| CN108056774A (en) * | 2017-12-29 | 2018-05-22 | 中国人民解放军战略支援部队信息工程大学 | Experimental paradigm mood analysis implementation method and its device based on visual transmission material |
| CN109034360A (en) * | 2018-07-13 | 2018-12-18 | 北京工业大学 | A kind of ant colony method constructing brain effective connectivity network from fMRI and DTI data |
| CN109222972A (en) * | 2018-09-11 | 2019-01-18 | 华南理工大学 | A kind of full brain data classification method of fMRI based on deep learning |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111539509A (en) * | 2020-04-16 | 2020-08-14 | 山东第一医科大学(山东省医学科学院) | Virtual digital brain construction method and system and intelligent robot control system |
| CN111539509B (en) * | 2020-04-16 | 2023-07-21 | 山东第一医科大学(山东省医学科学院) | Virtual digital brain construction method and system and intelligent robot control system |
| CN111861115A (en) * | 2020-06-15 | 2020-10-30 | 复旦大学 | Neural experiment test calculation method for consumer cognition of internet financial products |
| CN111861115B (en) * | 2020-06-15 | 2022-03-18 | 复旦大学 | Neural experiment test calculation method for consumer cognition of internet financial products |
| CN113177867A (en) * | 2021-04-22 | 2021-07-27 | 行云新能科技（深圳）有限公司 | Teaching method and device of chassis line control system and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110192860B (en) | 2022-10-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA2880758C (en) | Method and computing system for modelling a primate brain | |
| CN103800011B (en) | A kind of brain district based on functional mri effective connectivity analyzes system | |
| CN110169770A (en) | The fine granularity visualization system and method for mood brain electricity | |
| CN110192860A (en) | A kind of the Brian Imaging intelligent test analyzing method and system of network-oriented information cognition | |
| CN112465059A (en) | Multi-person motor imagery identification method based on cross-brain fusion decision and brain-computer system | |
| CN104361318A (en) | Disease diagnosis auxiliary system and disease diagnosis auxiliary method both based on diffusion tensor imaging technology | |
| CN109508755B (en) | Psychological assessment method based on image cognition | |
| CN111714118A (en) | Brain cognition model fusion method based on ensemble learning | |
| Li et al. | A novel method for the evaluation of fashion product design based on data mining | |
| CN109816630A (en) | FMRI visual coding model building method based on transfer learning | |
| CN110459317A (en) | Alzheimer disease assistant diagnosis system and method based on the dynamic brain network kernel of graph | |
| CN110717461A (en) | Fatigue state identification method, device and equipment | |
| CN109815855A (en) | A kind of electronic equipment automatic test approach and system based on machine learning | |
| CN110059232A (en) | A kind of data visualization method based on user experience measurement | |
| CN110889335B (en) | Human skeleton double interaction behavior identification method based on multichannel space-time fusion network | |
| CN106510736A (en) | Psychological state judging method and system based on multidimensional psychological state indexes | |
| Tu et al. | A state-space model for inferring effective connectivity of latent neural dynamics from simultaneous EEG/fMRI | |
| Kunanbayev et al. | Data augmentation for p300-based brain-computer interfaces using generative adversarial networks | |
| CN110473176A (en) | Image processing method and device, method for processing fundus images, electronic equipment | |
| CN117542121A (en) | Computer vision-based intelligent training and checking system and method | |
| CN111951950B (en) | Three-dimensional data medical classification system based on deep learning | |
| CN115953834A (en) | Multi-head attention posture estimation method and detection system for sit-up | |
| CN109146005A (en) | A kind of brain cognitive ability analysis method based on rarefaction representation | |
| CN111881979B (en) | Multi-modal data annotation device and computer-readable storage medium containing program | |
| CN114494132A (en) | Disease classification system based on deep learning and fiber bundle spatial statistical analysis |
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 |