CN109498037A - The brain cognitive measurement method of feature and multiple dimension-reduction algorithm is extracted based on deep learning - Google Patents
The brain cognitive measurement method of feature and multiple dimension-reduction algorithm is extracted based on deep learning Download PDFInfo
- Publication number
- CN109498037A CN109498037A CN201811573433.0A CN201811573433A CN109498037A CN 109498037 A CN109498037 A CN 109498037A CN 201811573433 A CN201811573433 A CN 201811573433A CN 109498037 A CN109498037 A CN 109498037A
- Authority
- CN
- China
- Prior art keywords
- feature
- brain
- local feature
- deep learning
- reduction algorithm
- 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
- 210000004556 brain Anatomy 0.000 title claims abstract description 68
- 230000001149 cognitive effect Effects 0.000 title claims abstract description 28
- 238000013135 deep learning Methods 0.000 title claims abstract description 22
- 238000000691 measurement method Methods 0.000 title claims abstract description 18
- 238000000034 method Methods 0.000 claims abstract description 44
- 230000003930 cognitive ability Effects 0.000 claims abstract description 30
- 238000000605 extraction Methods 0.000 claims abstract description 29
- 239000013598 vector Substances 0.000 claims abstract description 28
- 238000005259 measurement Methods 0.000 claims abstract description 20
- 230000009467 reduction Effects 0.000 claims abstract description 19
- 238000013527 convolutional neural network Methods 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 4
- 238000012706 support-vector machine Methods 0.000 claims description 4
- 230000019771 cognition Effects 0.000 claims 5
- 230000006870 function Effects 0.000 description 7
- 230000008901 benefit Effects 0.000 description 4
- 239000000284 extract Substances 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 238000000354 decomposition reaction Methods 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 210000004885 white matter Anatomy 0.000 description 3
- 230000032683 aging Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 210000003792 cranial nerve Anatomy 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000009792 diffusion process Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 210000004884 grey matter Anatomy 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 210000003484 anatomy Anatomy 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003925 brain function Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B2576/00—Medical imaging apparatus involving image processing or analysis
- A61B2576/02—Medical imaging apparatus involving image processing or analysis specially adapted for a particular organ or body part
- A61B2576/026—Medical imaging apparatus involving image processing or analysis specially adapted for a particular organ or body part for the brain
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Software Systems (AREA)
- Psychiatry (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Child & Adolescent Psychology (AREA)
- Developmental Disabilities (AREA)
- Educational Technology (AREA)
- Hospice & Palliative Care (AREA)
- Evolutionary Computation (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- Pathology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
- Image Analysis (AREA)
Abstract
The invention belongs to brain cognitive ability fields of measurement, more particularly to a kind of brain cognitive measurement method for extracting feature and multiple dimension-reduction algorithm based on deep learning, it is intended in order to solve the problems, such as the intelligence and facilitation of the high-dimensional extraction of brain image feature and cognitive ability measurement.The method of the present invention includes: the brain image data to input, the feature extraction of multichannel is carried out using feature extraction network, and obtain a local feature vectors by the way that concatenation is straightened;Rectangular projection dimensionality reduction is carried out to the local feature vectors of acquisition;Based on the cognitive ability constructed in advance-local feature corresponding relationship, the measurement of cognitive ability is carried out to the local feature after dimensionality reduction, and exports measurement result.The present invention realizes automation, intelligence and the facilitation of brain cognitive ability measurement;Recognition accuracy with higher simultaneously.
Description
Technical field
The invention belongs to brain cognitive ability fields of measurement, and in particular to a kind of to extract feature and multiple dimensionality reduction in deep learning
The brain cognitive measurement method of algorithm.
Background technique
The mode of measurement human brain cognitive ability at present, the mainly mode of questionnaire measurement, such as examination test, questionnaire survey
Etc. modes.This mode would generally be by the interference of supervisor's factor, the influence including main examination or the mood or the state of mind that are tested,
Therefore it is generally difficult to obtain and stablizes objective evaluation result.With the progress of cranial nerve image technology, have been able to high space-time
The form measurement of resolution ratio obtains the structure and function action message of brain.But the brain image data of this three-dimensional contains greatly
How the feature of the brain structure and function of amount measures the cognitive ability of human brain using these features, this is to conventional based on spy
The machine learning method of sign proposes huge challenge: firstly, the object identification being different from conventional computer vision is (as identified
Body form classification etc.), it is validity feature which feature can not be manually explicitly defined on brain image, in particular for functional neurosurgery
Image, artificial (including doctor) can not judging characteristic value;Secondly as three-dimensional brain image contains largely for a
The data of body, therefore the screening of feature just becomes particularly critical.The present invention proposes aiming at the two hierarchy of skill, quasi-solution
The human brain cognitive ability prediction for certainly automatically generating and screening for the feature of high-dimensional brain image.
Summary of the invention
In order to solve the above problem in the prior art, energy is extracted and recognized in order to solve high-dimensional brain image feature
The intelligence and facilitation problem, an aspect of of the present present invention that power measures propose a kind of based on deep learning extraction feature and more
The brain cognitive measurement method of weight dimension-reduction algorithm, comprising:
Step S10 carries out the feature extraction of multichannel using feature extraction network to the brain image data of input, and leads to
It crosses and concatenation one local feature vectors of acquisition is straightened;The feature extraction network is constructed based on convolutional neural networks;
Step S20 carries out rectangular projection dimensionality reduction to the local feature vectors obtained in step S10;
Step S30, based on the cognitive ability constructed in advance-local feature corresponding relationship, to the local feature after dimensionality reduction into
The measurement of row cognitive ability, and export measurement result;
Wherein, the cognitive ability-local feature corresponding relationship are as follows: based on local feature vectors sample and corresponding recognize
Know ability label, the mapping table of the cognitive ability and local feature that obtain by linear or nonlinear regression method
Show.
In some preferred embodiments, the local feature sample, acquisition methods are as follows:
By brain image data sample, local feature vectors are obtained by the method for step S10, and by the method for S20 into
Row dimensionality reduction obtains.
In some preferred embodiments, the brain image data is three-dimensional image.
In some preferred embodiments, the feature extraction network is the convolution mind comprising two or more convolutional layers
Through network.
In some preferred embodiments, the convolutional layer of the feature extraction network is three.
In some preferred embodiments, it in the cognitive ability-local feature corresponding relationship acquisition process, is used
Linear or nonlinear regression method be support vector machines linear or nonlinear regression method.
In some preferred embodiments, the cognitive ability-local feature corresponding relationship further includes preset confidence interval.
In some preferred embodiments, the feature extraction network, training sample include brain image data sample, part
Feature vector sample.
The second aspect of the present invention proposes a kind of storage device, wherein be stored with a plurality of program, described program be suitable for by
Processor is loaded and is executed to realize the above-mentioned brain cognitive measurement side for being extracted feature and multiple dimension-reduction algorithm based on deep learning
Method.
The third aspect of the present invention proposes a kind of processing unit, including processor, storage device;Processor, suitable for holding
Each program of row;Storage device is suitable for storing a plurality of program;Described program is suitable for being loaded by processor and being executed above-mentioned to realize
Based on deep learning extract feature and multiple dimension-reduction algorithm brain cognitive measurement method.
Beneficial effects of the present invention:
Feature extraction is carried out by feature extraction network through the invention, avoids and feature extraction manually is carried out to brain image
Intervene, and combine the cognitive ability-local feature corresponding relationship constructed based on mass data sample, realizes the survey of brain cognitive ability
Automation, intelligence and the facilitation of amount;Recognition accuracy with higher simultaneously.And it is right after multiple orthogonal projection algorithm
Magnanimity feature carries out dimensionality reduction so that can also maintain the validity of subsequent prediction while simplifying feature, improve stability and
Generalization Capability.
Detailed description of the invention
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 is the brain cognitive measurement that feature and multiple dimension-reduction algorithm are extracted based on deep learning of an embodiment of the present invention
Method flow schematic diagram.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to the embodiment of the present invention
In technical solution be clearly and completely described, it is clear that described embodiments are some of the embodiments of the present invention, without
It is whole embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not before making creative work
Every other embodiment obtained is put, shall fall within the protection scope of the present invention.
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
Convenient for description, part relevant to related invention is illustrated only in attached drawing.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.
A kind of brain cognitive measurement method for extracting feature and multiple dimension-reduction algorithm based on deep learning of the invention, such as Fig. 1
It is shown, comprising:
Step S10 carries out the feature extraction of multichannel using feature extraction network to the brain image data of input, and leads to
It crosses and concatenation one local feature vectors of acquisition is straightened;The feature extraction network is constructed based on convolutional neural networks;
Step S20 carries out rectangular projection dimensionality reduction to the local feature vectors obtained in step S10;
Step S30, based on the cognitive ability constructed in advance-local feature corresponding relationship, to the local feature after dimensionality reduction into
The measurement of row cognitive ability, and export measurement result;
Wherein, the cognitive ability-local feature corresponding relationship are as follows: based on local feature vectors sample and corresponding recognize
Know ability label, the mapping table of the cognitive ability and local feature that obtain by linear or nonlinear regression method
Show.
In order to more clearly to the present invention is based on the brain cognitive measurement sides that deep learning extracts feature and multiple dimension-reduction algorithm
Method is illustrated, and carries out expansion detailed description to each step in a kind of embodiment of our inventive method with reference to the accompanying drawing.
Step S10 carries out the feature extraction of multichannel using feature extraction network to the brain image data of input, and leads to
It crosses and concatenation one local feature vectors of acquisition is straightened;Feature extraction network is constructed based on convolutional neural networks.
The brain image data that user inputs in the embodiment is usually three-dimensional brain image, such as magnetic resonance T1, T2 image
Deng;For Brain mapping picture or diffusion magnetic resonance image, it is also possible to the dimension including time dimension either Diffusion direction
Degree;For the above-mentioned brain image more than 3 dimensions, in cranial nerve image field, it will usually calculate the mark for being directed to each voxel
Amount indicates, such as the brain function joint efficiency of voxel level, white matter fiber structure myelinization degree coefficient etc..Therefore, although
Theoretically, method of the invention can be supported to consider computer hardware more than three-dimensional image input, but in actually calculating
Limitation of computing capability, such as the limitation of GPU video memory etc., it will usually which being converted directly into more can directly characterize brain structure and function
The three-dimensional image of mobility.
The basic ideas of used convolutional neural networks (Convolutional Neural Network, CNN) are same
When by a variety of convolution collecting images carry out convolution, then chosen in each convolution region most representative voxel (namely
It is pondization operation, pooling), this process is repeated (according to user computer hardware using multilayer neural network in each channel
Performance, the convolution number of plies can choose two layers or more numbers of plies;It is found by test, complexity can calculated using 3 layer networks
It is achieved a better balance between degree and precision).For each filtering image channel, Preliminary design of the present invention is based on three-dimensional figure
The multilayer convolutional neural networks of picture have separately included convolution (convolution) He Chihua (pooling) behaviour for brain image
Make, and by comparing, pondization operation selects maximum pond (Maxpooling) more more effective than other mean value ponds etc..Entire system
System input is original three-dimensional image (structural images or function image), final network output be multiple channels size compared with
Small 3-D image, each image represent the channel filter generate to the maximally related feature of predictive variable, the series three
Dimension image directly can be spliced into a column vector by the way that concatenation (flattern) is straightened, and using the column vector as part
Feature vector.Local feature vectors are to constitute after multilayer convolution sum pondization operates with the highest feature of the predictive variable degree of correlation
Vector, such as Pearson correlation coefficient, this feature are the comprehensive contribution of many adjacent voxels, usually not fixed anatomy meaning
Justice.
Step S20 carries out rectangular projection dimensionality reduction to the local feature vectors obtained in step S10.
Projection matrix U and V are obtained based on sample database: assuming that the brain image data sample in sample database is through step
The collection of feature after S10 extraction is combined into Xm×n, (m is number of samples, the number that n is characterized), cognitive ability mark in sample database
Label are Ym×k, k≤n, to Xm×n、Ym×kIt carries out comprehensive rectangular projection using the method for partial correlation least square to decompose, such as formula (1)
It is shown:
R=YTX=U Δ VT (1)
Wherein, R indicates the covariance matrix of Y and X, U Δ VTBy to YTX is obtained after carrying out singular value decomposition, and then is obtained
Two orthogonal matrixes U and V.
Therefore, the projection L of the space available X and Y after orthogonal transformationXAnd LY, respectively as shown in formula (2), (3),
LX=XV (2)
LY=YU (3)
Because rectangular projection is related to multiple brain areas, the feature finally selected may be the feature of certain single brain area,
It is likely to be the combination of the feature of multiple brain areas.Therefore, it is to the local feature obtained after the local feature dimensionality reduction of grey matter, white matter
The combination of the feature of the grey matter of certain brain areas/white matter feature or multiple brain areas.
In practical calculating process, the mode of dimensionality reduction can be realized by choosing main component.Using passing through sample number
According to projection matrix U and V that library obtains, sample is tested to newcomer, after being extracted local feature in the manner described above, according to
Projection matrix U and V carries out Projective decomposition dimensionality reduction using the method for formula (2), (3) to the local feature of measurand.This method
Reduce the redundancy of variable not only through rectangular projection, and considers independent variable and dependent variable while Orthogonal Decomposition
Common spatial information so that can more effectively remove it is subsequent return information redundancy.This be with traditional dimensionality reduction mode, such as
Maximum advantage is compared in the methods of principal component analysis (PCA).
Step S30, based on the cognitive ability constructed in advance-local feature corresponding relationship, to the local feature after dimensionality reduction into
The measurement of row cognitive ability, and export measurement result.
It, can be to the office that input brain image data extracts based on the cognitive ability constructed in advance-local feature corresponding relationship
Portion's feature vector carries out cognitive ability acquisition and increases sample in cognitive ability-local feature corresponding relationship in some embodiments
The factor of notebook data actual age corresponding to input brain image data can be tested main body and carry out cognitive ability comparison of the same age, lead to
Increase confidence interval is crossed, to judge that cognitive ability is normal or abnormal, then indicates that cognitive ability belongs in section for example, falling into and writing
Conventional levels then indicate to be apparently higher than average level in the top of confidence interval, then indicate obvious low in the lower section of confidence interval
In average level.So as to be used for development or ageing level research, to judge to be in brain area local horizontal and full brain level
Hypoevolutism or aging are too fast.
Cognitive ability-local feature corresponding relationship is building in advance in the present embodiment, based on the original of great amount of samples data
Sample data set obtains local feature vectors by the method for step S10, and carries out dimensionality reduction by the method for S20 and obtain part
Feature vector sample constructs the second sample data set.Original sample includes brain image data sample, cognitive ability label, may be used also
Including age label;Corresponding second sample data includes local feature vectors sample, cognitive ability label, it is corresponding can be with
Including age label.Based on the second sample data set, the cognitive ability that is obtained by linear or nonlinear regression method and
The corresponding relationship of local feature indicates.It is preferred that linear or nonlinear regression the method using support vector machines (SVM) carries out
The building of the corresponding relationship of cognitive ability and local feature, the great advantage of this method in systems be the vector after dimensionality reduction again
It is secondary to be transmitted to higher dimensional space to enhance classifying distance.
Training sample used in the training of feature extraction network is by brain image data sample, local feature vectors in the present embodiment
Sample is constituted, and feature extraction network is trained and is optimized by the training sample set for including a large amount of training samples, is being used
In the process, local feature vectors are carried out using brain image data of the trained feature extraction network to the measurand of input
It extracts.
A kind of storage device of second embodiment of the invention, wherein being stored with a plurality of program, described program is suitable for by handling
Device is loaded and is executed to realize the above-mentioned brain cognitive measurement method for being extracted feature and multiple dimension-reduction algorithm based on deep learning.
A kind of processing unit of third embodiment of the invention, including processor, storage device;Processor is adapted for carrying out each
Program;Storage device is suitable for storing a plurality of program;Described program is suitable for being loaded by processor and being executed to realize above-mentioned base
The brain cognitive measurement method of feature and multiple dimension-reduction algorithm is extracted in deep learning.
Person of ordinary skill in the field can be understood that, for convenience and simplicity of description, foregoing description
The specific work process and related explanation of storage device, processing unit, can refer to corresponding processes in the foregoing method embodiment,
Details are not described herein.
Those skilled in the art should be able to recognize that, side described in conjunction with the examples disclosed in the embodiments of the present disclosure
Method step, can be realized with electronic hardware, computer software, or a combination of the two, and software module, method and step are corresponding
Program can be placed in random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable
Any other form of storage medium well known in ROM, register, hard disk, moveable magnetic disc, CD-ROM or technical field
In.In order to clearly demonstrate the interchangeability of electronic hardware and software, generally retouched according to function in the above description
Each exemplary composition and step are stated.These functions are executed actually with electronic hardware or software mode, depend on technical side
The specific application and design constraint of case.Those skilled in the art can carry out each specific application to come using distinct methods
Realize described function, but such implementation should not be considered as beyond the scope of the present invention.
Term " first ", " second " etc. are to be used to distinguish similar objects, rather than be used to describe or indicate specific suitable
Sequence or precedence.
Term " includes " or any other like term are intended to cover non-exclusive inclusion, so that including a system
Process, method, article or equipment/device of column element not only includes those elements, but also including being not explicitly listed
Other elements, or further include the intrinsic element of these process, method, article or equipment/devices.
So far, it has been combined preferred embodiment shown in the drawings and describes technical solution of the present invention, still, this field
Technical staff is it is easily understood that protection scope of the present invention is expressly not limited to these specific embodiments.Without departing from this
Under the premise of the principle of invention, those skilled in the art can make equivalent change or replacement to the relevant technologies feature, these
Technical solution after change or replacement will fall within the scope of protection of the present invention.
Claims (10)
1. a kind of brain cognitive measurement method for extracting feature and multiple dimension-reduction algorithm based on deep learning characterized by comprising
Step S10 carries out the feature extraction of multichannel using feature extraction network to the brain image data of input, and passes through drawing
Straight concatenation obtains a local feature vectors;The feature extraction network is constructed based on convolutional neural networks;
Step S20 carries out rectangular projection dimensionality reduction to the local feature vectors obtained in step S10;
Step S30 recognizes the local feature after dimensionality reduction based on the cognitive ability constructed in advance-local feature corresponding relationship
Know the measurement of ability, and exports measurement result;
Wherein, the cognitive ability-local feature corresponding relationship are as follows: be based on local feature vectors sample and corresponding cognition energy
The corresponding relationship of power label, the cognitive ability and local feature that are obtained by linear or nonlinear regression method indicates.
2. the brain cognitive measurement method according to claim 1 that feature and multiple dimension-reduction algorithm are extracted based on deep learning,
It is characterized in that, the local feature sample, acquisition methods are as follows:
By brain image data sample, local feature vectors are obtained by the method for step S10, and dropped by the method for S20
Dimension obtains.
3. the brain cognitive measurement method according to claim 1 that feature and multiple dimension-reduction algorithm are extracted based on deep learning,
It is characterized in that, the brain image data is three-dimensional image.
4. the brain cognitive measurement method according to claim 1 that feature and multiple dimension-reduction algorithm are extracted based on deep learning,
It is characterized in that, the feature extraction network is the convolutional neural networks comprising two or more convolutional layers.
5. the brain cognitive measurement method according to claim 3 that feature and multiple dimension-reduction algorithm are extracted based on deep learning,
It is characterized in that, the convolutional layer of the feature extraction network is three.
6. the brain cognition according to any one of claims 1-5 for extracting feature and multiple dimension-reduction algorithm based on deep learning
Measurement method, which is characterized in that used linear in the cognitive ability-local feature corresponding relationship acquisition process
Either the method for nonlinear regression is linear or nonlinear regression the method for support vector machines.
7. the brain cognition according to any one of claims 1-5 for extracting feature and multiple dimension-reduction algorithm based on deep learning
Measurement method, which is characterized in that the cognitive ability-local feature corresponding relationship further includes preset confidence interval.
8. the brain cognition according to any one of claims 1-5 for extracting feature and multiple dimension-reduction algorithm based on deep learning
Measurement method, which is characterized in that the feature extraction network, training sample include brain image data sample, local feature to
Measure sample.
9. a kind of storage device, wherein being stored with a plurality of program, which is characterized in that described program is suitable for being loaded and being held by processor
Row is to realize that the brain cognition of any of claims 1-8 for extracting feature and multiple dimension-reduction algorithm based on deep learning is surveyed
Amount method.
10. a kind of processing unit, including processor, storage device;Processor is adapted for carrying out each program;Storage device is suitable for
Store a plurality of program;It is characterized in that, described program is any in claim 1-8 to realize suitable for being loaded by processor and being executed
The brain cognitive measurement method that feature and multiple dimension-reduction algorithm are extracted based on deep learning described in.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811573433.0A CN109498037B (en) | 2018-12-21 | 2018-12-21 | Brain cognition measurement method based on deep learning extraction features and multiple dimension reduction algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811573433.0A CN109498037B (en) | 2018-12-21 | 2018-12-21 | Brain cognition measurement method based on deep learning extraction features and multiple dimension reduction algorithm |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109498037A true CN109498037A (en) | 2019-03-22 |
CN109498037B CN109498037B (en) | 2020-06-16 |
Family
ID=65754581
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811573433.0A Active CN109498037B (en) | 2018-12-21 | 2018-12-21 | Brain cognition measurement method based on deep learning extraction features and multiple dimension reduction algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109498037B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109646001A (en) * | 2018-12-21 | 2019-04-19 | 中国科学院自动化研究所 | Brain cognitive ability measuring system based on brain image |
CN111728590A (en) * | 2020-06-30 | 2020-10-02 | 中国人民解放军国防科技大学 | Individual cognitive ability prediction method and system based on dynamic function connection |
CN111839506A (en) * | 2019-04-30 | 2020-10-30 | 清华大学 | Mental load detection method and device |
CN113393141A (en) * | 2021-06-23 | 2021-09-14 | 科大讯飞股份有限公司 | Reading capability evaluation method, device and equipment |
CN113408852A (en) * | 2021-05-18 | 2021-09-17 | 江西师范大学 | Meta-cognition ability evaluation model based on online learning behavior and deep neural network |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102156995A (en) * | 2011-04-21 | 2011-08-17 | 北京理工大学 | Video movement foreground dividing method in moving camera |
CN102663414A (en) * | 2012-03-14 | 2012-09-12 | 大连灵动科技发展有限公司 | Cerebrum cognitive status recognition method based on cerebral function imaging |
CN103116764A (en) * | 2013-03-02 | 2013-05-22 | 西安电子科技大学 | Brain cognitive state judgment method based on polyteny principal component analysis |
CN103440512A (en) * | 2013-09-17 | 2013-12-11 | 西安电子科技大学 | Identifying method of brain cognitive states based on tensor locality preserving projection |
CN104951763A (en) * | 2015-06-16 | 2015-09-30 | 北京四方继保自动化股份有限公司 | Power generator set subsynchronous risk evaluating method based on wave recording big data abnormal detection |
CN105046709A (en) * | 2015-07-14 | 2015-11-11 | 华南理工大学 | Nuclear magnetic resonance imaging based brain age analysis method |
CN105654138A (en) * | 2015-12-30 | 2016-06-08 | 广东工业大学 | Orthogonal projection and dimensionality reduction classification method and system for multidimensional data |
CN106096616A (en) * | 2016-06-08 | 2016-11-09 | 四川大学华西医院 | A kind of nuclear magnetic resonance image feature extraction based on degree of depth study and sorting technique |
CN106156484A (en) * | 2016-06-08 | 2016-11-23 | 中国科学院自动化研究所 | Disease of brain individuation Forecasting Methodology based on nuclear magnetic resonance image and system |
CN106846316A (en) * | 2017-02-10 | 2017-06-13 | 云南电网有限责任公司电力科学研究院 | A kind of GIS inside typical defect automatic distinguishing method for image |
US20180204046A1 (en) * | 2015-08-04 | 2018-07-19 | Siemens Aktiengesellschaft | Visual representation learning for brain tumor classification |
-
2018
- 2018-12-21 CN CN201811573433.0A patent/CN109498037B/en active Active
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102156995A (en) * | 2011-04-21 | 2011-08-17 | 北京理工大学 | Video movement foreground dividing method in moving camera |
CN102663414A (en) * | 2012-03-14 | 2012-09-12 | 大连灵动科技发展有限公司 | Cerebrum cognitive status recognition method based on cerebral function imaging |
CN103116764A (en) * | 2013-03-02 | 2013-05-22 | 西安电子科技大学 | Brain cognitive state judgment method based on polyteny principal component analysis |
CN103440512A (en) * | 2013-09-17 | 2013-12-11 | 西安电子科技大学 | Identifying method of brain cognitive states based on tensor locality preserving projection |
CN104951763A (en) * | 2015-06-16 | 2015-09-30 | 北京四方继保自动化股份有限公司 | Power generator set subsynchronous risk evaluating method based on wave recording big data abnormal detection |
CN105046709A (en) * | 2015-07-14 | 2015-11-11 | 华南理工大学 | Nuclear magnetic resonance imaging based brain age analysis method |
US20180204046A1 (en) * | 2015-08-04 | 2018-07-19 | Siemens Aktiengesellschaft | Visual representation learning for brain tumor classification |
CN105654138A (en) * | 2015-12-30 | 2016-06-08 | 广东工业大学 | Orthogonal projection and dimensionality reduction classification method and system for multidimensional data |
CN106096616A (en) * | 2016-06-08 | 2016-11-09 | 四川大学华西医院 | A kind of nuclear magnetic resonance image feature extraction based on degree of depth study and sorting technique |
CN106156484A (en) * | 2016-06-08 | 2016-11-23 | 中国科学院自动化研究所 | Disease of brain individuation Forecasting Methodology based on nuclear magnetic resonance image and system |
CN106846316A (en) * | 2017-02-10 | 2017-06-13 | 云南电网有限责任公司电力科学研究院 | A kind of GIS inside typical defect automatic distinguishing method for image |
Non-Patent Citations (1)
Title |
---|
崔兴伟: "基于功能磁共振成像的Ⅱ型糖尿病患者认知功能预测分析", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109646001A (en) * | 2018-12-21 | 2019-04-19 | 中国科学院自动化研究所 | Brain cognitive ability measuring system based on brain image |
CN111839506A (en) * | 2019-04-30 | 2020-10-30 | 清华大学 | Mental load detection method and device |
CN111728590A (en) * | 2020-06-30 | 2020-10-02 | 中国人民解放军国防科技大学 | Individual cognitive ability prediction method and system based on dynamic function connection |
CN113408852A (en) * | 2021-05-18 | 2021-09-17 | 江西师范大学 | Meta-cognition ability evaluation model based on online learning behavior and deep neural network |
CN113408852B (en) * | 2021-05-18 | 2022-04-19 | 江西师范大学 | Meta-cognition ability evaluation model based on online learning behavior and deep neural network |
CN113393141A (en) * | 2021-06-23 | 2021-09-14 | 科大讯飞股份有限公司 | Reading capability evaluation method, device and equipment |
CN113393141B (en) * | 2021-06-23 | 2023-12-01 | 科大讯飞股份有限公司 | Method, device and equipment for evaluating reading capability |
Also Published As
Publication number | Publication date |
---|---|
CN109498037B (en) | 2020-06-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109498037A (en) | The brain cognitive measurement method of feature and multiple dimension-reduction algorithm is extracted based on deep learning | |
CN110532920B (en) | Face recognition method for small-quantity data set based on FaceNet method | |
CN110659665B (en) | Model construction method of different-dimension characteristics and image recognition method and device | |
CN106960210B (en) | The method and apparatus of target detection | |
Ahishali et al. | Advance warning methodologies for covid-19 using chest x-ray images | |
CN109800811A (en) | A kind of small sample image-recognizing method based on deep learning | |
CN109726746A (en) | A kind of method and device of template matching | |
CA2804439A1 (en) | System and method for categorizing an image | |
CN111134664B (en) | Epileptic discharge identification method and system based on capsule network and storage medium | |
Leeds et al. | Comparing visual representations across human fMRI and computational vision | |
CN104281835A (en) | Face recognition method based on local sensitive kernel sparse representation | |
Köster et al. | A two-layer model of natural stimuli estimated with score matching | |
CN110532950A (en) | Video feature extraction method, micro- expression recognition method based on micro- expression video | |
CN111680579A (en) | Remote sensing image classification method for adaptive weight multi-view metric learning | |
CN109978074A (en) | Image aesthetic feeling and emotion joint classification method and system based on depth multi-task learning | |
Mermillod et al. | The coarse-to-fine hypothesis revisited: Evidence from neuro-computational modeling | |
Ritchie et al. | A varying role for abstraction in models of category learning constructed from neural representations in early visual cortex | |
Dhinagar et al. | Efficiently Training Vision Transformers on Structural MRI Scans for Alzheimer’s Disease Detection | |
Manzo et al. | FastGCN+ ARSRGemb: a novel framework for object recognition | |
Zhang et al. | Saliency detection via sparse reconstruction and joint label inference in multiple features | |
CN115392474B (en) | Local perception graph representation learning method based on iterative optimization | |
CN116311504A (en) | Small sample behavior recognition method, system and equipment | |
Xia et al. | Media quality assessment by perceptual gaze-shift patterns discovery | |
CN115205877A (en) | Irregular typesetting invoice document layout prediction method and device and storage medium | |
CN108256569A (en) | A kind of object identifying method under complex background and the computer technology used |
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 |