AU2021102810A4 - A system for human cognitive states classification - Google Patents
A system for human cognitive states classification Download PDFInfo
- Publication number
- AU2021102810A4 AU2021102810A4 AU2021102810A AU2021102810A AU2021102810A4 AU 2021102810 A4 AU2021102810 A4 AU 2021102810A4 AU 2021102810 A AU2021102810 A AU 2021102810A AU 2021102810 A AU2021102810 A AU 2021102810A AU 2021102810 A4 AU2021102810 A4 AU 2021102810A4
- Authority
- AU
- Australia
- Prior art keywords
- classification
- fmri
- data
- dataset
- voxels
- 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.)
- Ceased
Links
- 230000006998 cognitive state Effects 0.000 title claims abstract description 17
- 238000002599 functional magnetic resonance imaging Methods 0.000 claims abstract description 67
- 238000000034 method Methods 0.000 claims abstract description 47
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 32
- 230000002068 genetic effect Effects 0.000 claims abstract description 22
- 238000012549 training Methods 0.000 claims description 18
- 238000012360 testing method Methods 0.000 claims description 16
- 238000013480 data collection Methods 0.000 claims description 13
- 230000035772 mutation Effects 0.000 claims description 8
- 210000000349 chromosome Anatomy 0.000 claims description 7
- 230000007423 decrease Effects 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims 1
- 230000035945 sensitivity Effects 0.000 abstract description 5
- 238000011160 research Methods 0.000 abstract description 3
- 210000004556 brain Anatomy 0.000 description 14
- 230000008901 benefit Effects 0.000 description 9
- 238000013459 approach Methods 0.000 description 8
- 238000012880 independent component analysis Methods 0.000 description 6
- 238000003062 neural network model Methods 0.000 description 6
- 230000000284 resting effect Effects 0.000 description 6
- 238000005457 optimization Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 239000008280 blood Substances 0.000 description 2
- 210000004369 blood Anatomy 0.000 description 2
- 230000007177 brain activity Effects 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000006213 oxygenation reaction Methods 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000007637 random forest analysis Methods 0.000 description 2
- INGWEZCOABYORO-UHFFFAOYSA-N 2-(furan-2-yl)-7-methyl-1h-1,8-naphthyridin-4-one Chemical compound N=1C2=NC(C)=CC=C2C(O)=CC=1C1=CC=CO1 INGWEZCOABYORO-UHFFFAOYSA-N 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 108010002255 deoxyhemoglobin Proteins 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000002595 magnetic resonance imaging Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012421 spiking Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/211—Selection of the most significant subset of features
- G06F18/2111—Selection of the most significant subset of features by using evolutionary computational techniques, e.g. genetic algorithms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Public Health (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- Quality & Reliability (AREA)
- Primary Health Care (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Computing Systems (AREA)
- Physiology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present disclosure relates to a system for human cognitive states classification. In the
present disclosure a new system is developed to determine the cognitive state of the fMRI data
from Star Plus and Hax by datasets. Homogeneity of fMRI data is calculated by k-NN based
genetic algorithm. If the distance between the voxels is small, it will be high degree of similarity
and it is classified by GSW-LCDRC in order to obtain better classification accuracy. The GSW
LCDRC reduces the "curse of dimensionality" problem. Finally, the proposed methodology
performance was validated by comparing the previous research works in terms of sensitivity,
specificity, accuracy, overall sensitivity, overall specificity and overall accuracy.
13
m
0Q
zM
U
q = u.u
p - - 4
UC
IF~
Description
m 0Q
zM U
q = u.u p - - 4 UC
IF~
The present disclosure relates to a system for human cognitive states classification.
Currently, fMRI is the most emerging technique, which is used for examining the human brain activity. The fMRI technique utilizes MRI for calculating the oxygenation blood changes related to the activity of brain. The most common mechanism in fMRI technique is blood oxygenation level-dependent (BOLD) to understand the changes in deoxy-hemoglobin concentration in a brain. In fMRI, infer the brain activity using BOLD signal for generating the three-dimensional images and each image stores the BOLD signal information by the mean of voxels.
Analyzing the fMRI data using the human exerts is a time consuming process and also it is not reliable, so for this machine learning based methodologies have been developed. In recent fMRI studies these methodologies plays an important and essential role in extracting the useful patterns and information from large fMRI datasets with missing and noise values.
In fMRI feature selection plays an important role in identifying the optimal voxel subset, this is why it is important to select the suitable feature selection approach to enhances the classification accuracy regardless of the sample size. For this several selection methodologies are implemented in fMRI application. After selecting the optimal voxels, classification is carried out for pattern classification.
In one existing solution an elastic net approach has been developed for feature selection in order to determine the voxels and SVM was used for voxels classification. Elastic net generates a penalty function (either ridge or lasso penalty) for voxel optimization. Selecting the penalty function was difficult in elastic net methodology.
In another existing solution random forest and independent component analysis (ICA) have been presented for fMRI data classification. Dimensionality reduction was done by using ICA, and then classification was carried out by random forest. Here, ICA depends on the assumption of the signal source (independent components), and violation in the assumption degrades the performance.
In another existing solution particle swarm optimization (PSO) and SVM have been developed for FMRI data classification. In this disclosure feature selection was done by using PSO and SVM utilized for voxel classification. The major drawbacks of the PSO approach in fMRI classification were: low convergence rate and easily falls into high-dimensional space. In another existing solution ICA was used for feature selection and fMRI voxels were classified by using SVM. In ICA, it was difficult to select the number of threshold values and independent components for mapping in fMRI. In another existing solution the collected fMRI data has been preprocessed by using realignment smoothing. Then, feature selection was carried out by employing spatial demeaning and classifies the voxels by using SVM. In another existing solution scalar h-map was used for feature selection and SVM was utilized for classification. Here, SVM was a binary classifier, which supports only two-class classification.
In another existing solution rough set-based optimization and K-means approach were used for fMRI classification. The high-dimensional data affect the performance of the k-means approach, because it was difficult to predict the k-value. In another existing solution Widrow Hoff learning parallel LCDRC was developed to find the discriminant subspace by increasing the collaborative BCRE and decreasing the WCRE that helps to reduce the authentication error. While classifying a number of classes, the classification accuracy of the developed methodology highly degrades due to increase in complexity of the system. In another existing solution a deep learning-based adaptive LCDRC was developed for face recognition in order to improve the biometric authentication. The major drawback of deep learning-based adaptive LCDRC was more complex for identifying the free space projections, which can able to deliver better discriminant ability.
In one prior art solution the invention discloses an fMRI whole brain data classification method based on depth learning. The method comprises: (1) obtaining fMRI test data, preprocessing and obtaining corresponding labels; (2) aggregation of fMRI whole brain data; (3) slicing the average three-dimensional images in the orthogonal x, y and z axes; (4) converting three groups of two-dimensional images into one frame of multi-channel two-dimensional images respectively; (5) constructing a hybrid multi-channel convolution neural network model for fMRI whole brain data classification; (6) processing the fMRI data, training the obtained classification tags as input data, and using the obtained parameters in the hybrid convolution neural network model of fMRI whole brain data classification; (7) processing the fMRI data, and inputting the three multi-channel images into the trained mixed convolution neural network model for classification. The method can effectively improve the accuracy rate of the fMRI whole brain data classification, and simultaneously reduce the calculation amount of training and classification of the fMRI whole brain data classification model.
In another prior art solution the invention discloses a resting state fMRI data classification method and device based on deep learning. The method comprises: 1) acquiring resting state fMRI test data and performing preprocessing and obtaining tags; 2) performing brain region division on the resting state fMRI data, and extracting functional connectivity features and brain region comprehensive features; 3) extracting whole brain voxel point features; 4) extracting personal attribute features; 5) constructing a hybrid neural network model for resting state fMRI data classification; 6) processing the data for the model training part, subjecting the data as input data to mixed neural network training, and using the obtained parameters for the mixed neural network model of the resting state fMRI data classification; and 7) processing the resting state fMRI data, and inputting the obtained function connectivity features, brain region comprehensive features, whole brain voxel point features and personal attribute features into the trained hybrid neural network model for classification. The invention can retain the data form of each feature, comprehensively consider the information of each feature, and effectively improve the classification accuracy rate.
However, no existing research study has used fMRI data for distinguishing more than two classes. The high-dimensional features make fMRI data difficult for mining and classification, because if the volume of the data space increases, then the acquired data become sparse, which leads to the "curse of dimensionality" problem. Therefore in order to avoid the aforementioned drawbacks there is a need for a system for human cognitive states classification.
The present disclosure relates to a system for human cognitive states classification. The present disclosure proposes a new feature selection and classification methodology for classifying the human cognitive states from fMRI data. Initially, the fMRI data were collected from the Star Plus and Hax by datasets. Then, k-nearest neighbors algorithm (k-NN)-based genetic algorithm was developed to choose the optimal voxels from the active region of interests. The proposed approach selects the data to feature subsets based on k-NN algorithm, so the data volume was effectively reduced and the voxel information was maintained significantly. The most informative voxels were given as the input for gradient self-weighting that produces an optimal weight value. Respective weight value was added to the projection matrix of linear collaborative discriminant regression classification for identifying the future projection matrix that reduces the error between two individual voxels in subspace. The experimental outcome shows that the proposed methodology improved the accuracy in fMRI data classification up to 0.7-23% compared to the existing methods.
The present disclosure seeks to provide a system for human cognitive states classification. The system comprises: a data collection module for collecting at least a dataset, wherein the dataset comprises of a Functional magnetic resonance imaging (fMRI) data arranged sequentially; a feature selection module connected to the data collection module for identifying a relevant subset of fMRI data based on a particular criterion using a K-NN based genetic algorithm technique, wherein the genetic algorithm decreases redundancy within an input voxel and also determines the maximum relevance between an output and the input voxels, wherein the input voxel is calculated using a cross over and mutation operator; and a classification module connected to the feature selection module for classifying the obtained fMRI data using a Linear Collaborative Discriminant Regression Classification (LCDRC) upon dividing the dataset into a training set and a testing set.
The present disclosure also seeks to provide a method for human cognitive states classification. The method comprises: collecting at least a dataset using a data collection module, wherein the dataset comprises of a Functional magnetic resonance imaging (fMRI) data arranged sequentially; identifying a relevant subset of fMRI data using a feature selection module connected to the data collection module based on a particular criterion using a K-NN based genetic algorithm technique, wherein the genetic algorithm decreases redundancy within an input voxel and also determines the maximum relevance between an output and the input voxels, wherein the input voxel is calculated using a crossover and mutation operator; and classifying the obtained fMRI data using a Linear Collaborative Discriminant Regression Classification (LCDRC) of a classification module connected to the feature selection module upon dividing the dataset into a training set and a testing set.
An objective of the present disclosure is to provide system and method for human cognitive states classification.
Another object of the present disclosure is to collect the fMRI data from StarPlus and Hax by dataset.
Another object of the present disclosure is to develop k-nearest neighbors algorithm (k NN)-based genetic algorithm to choose the optimal voxels from the active region of interests.
Another object of the present disclosure is to preserves the fMRI data characteristics for interpretability using the proposed feature selection algorithm.
Another object of the present disclosure is to use GSW-LCDRC for classifying the high degree of similarity voxels from k-NN-based genetic algorithm in order to obtain better classification accuracy.
Yet, another object of the present disclosure is to reduce the "curse of dimensionality" problem using GSW-LCDRC by decreasing the within-class reconstruction error (WCRE) and increasing the between-class reconstruction error (BCRE) in the class space.
To further clarify advantages and features of the present disclosure, a more particular description of the invention will be rendered by reference to specific embodiments thereof, which is illustrated in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail with the accompanying drawings.
These and other features, aspects, and advantages of the present disclosure will become better understood when the following detailed description is read with reference to the accompanying drawings in which like characters represent like parts throughout the drawings, wherein:
Figure 1 illustrates a block diagram of a system for human cognitive states classification in accordance with an embodiment of the present disclosure;
Figure 2 illustrates a flow chart of a method for human cognitive states classification in accordance with an embodiment of the present disclosure;
Figure 3 illustrates the description of k-NN-based genetic algorithm in accordance with an embodiment of the present disclosure;
Figure 4 illustrates a table of Performance comparison of the existing and proposed work in accordance with an embodiment of the present disclosure;
Further, skilled artisans will appreciate that elements in the drawings are illustrated for simplicity and may not have been necessarily been drawn to scale. For example, the flow charts illustrate the method in terms of the most prominent steps involved to help to improve understanding of aspects of the present disclosure. Furthermore, in terms of the construction of the device, one or more components of the device may have been represented in the drawings by conventional symbols, and the drawings may show only those specific details that are pertinent to understanding the embodiments of the present disclosure so as not to obscure the drawings with details that will be readily apparent to those of ordinary skill in the art having benefit of the description herein.
For the purpose of promoting an understanding of the principles of the invention, reference will now be made to the embodiment illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the invention is thereby intended, such alterations and further modifications in the illustrated system, and such further applications of the principles of the invention as illustrated therein being contemplated as would normally occur to one skilled in the art to which the invention relates.
It will be understood by those skilled in the art that the foregoing general description and the following detailed description are exemplary and explanatory of the invention and are not intended to be restrictive thereof.
Reference throughout this specification to "an aspect", "another aspect" or similar language means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present disclosure. Thus, appearances of the phrase "in an embodiment", "in another embodiment" and similar language throughout this specification may, but do not necessarily, all refer to the same embodiment.
The terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process or method that comprises a list of steps does not include only those steps but may include other steps not expressly listed or inherent to such process or method. Similarly, one or more devices or sub-systems or elements or structures or components proceeded by "comprises...a" does not, without more constraints, preclude the existence of other devices or other sub-systems or other elements or other structures or other components or additional devices or additional sub-systems or additional elements or additional structures or additional components.
Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The system, methods, and examples provided herein are illustrative only and not intended to be limiting.
Embodiments of the present disclosure will be described below in detail with reference to the accompanying drawings.
Figure 1 illustrates a block diagram of a system for human cognitive states classification in accordance with an embodiment of the present disclosure. The system 100 includes a data collection module 102 for collecting at least a dataset, wherein the dataset comprises of a Functional magnetic resonance imaging (fMRI) data arranged sequentially.
In an embodiment a feature selection module 104 is connected to the data collection module for identifying a relevant subset of fMRI data based on a particular criterion using a K NN based genetic algorithm technique, wherein the genetic algorithm decreases redundancy within an input voxel and also determines the maximum relevance between an output and the input voxels, wherein the input voxel is calculated using a crossover and mutation operator. The feature selection module comprises: Generating an initial population of set as the subset of input voxels; identifying a discriminative capability of each subset of the voxels using a fitness evaluator; calculating fitness of each voxels using the K-NN algorithm technique; and finding a shortest distance between the training and the testing set for classification, wherein the k-NN algorithm uses a Euclidean distance measure for identifying the nearest distance between the training and testing sets for reducing the classification error.
In an embodiment a classification module 106 is connected to the feature selection module for classifying the obtained fMRI data using a Linear Collaborative Discriminant Regression Classification (LCDRC) upon dividing the dataset into a training set and a testing set.
Figure 2 illustrates a flow chart of a method for human cognitive states classification in accordance with an embodiment of the present disclosure. At step 202 the method 200 includes collecting at least a dataset using a data collection module, wherein the dataset comprises of a Functional magnetic resonance imaging (fMRI) data arranged sequentially.
At step 204 the method 200 includes identifying a relevant subset of fMRI data using a feature selection module connected to the data collection module based on a particular criterion using a K-NN based genetic algorithm technique, wherein the genetic algorithm decreases redundancy within an input voxel and also determines the maximum relevance between an output and the input voxels, wherein the input voxel is calculated using a crossover and mutation operator.
At step 206 the method 200 includes classifying the obtained fMRI data using a Linear Collaborative Discriminant Regression Classification (LCDRC) of a classification module connected to the feature selection module upon dividing the dataset into a training set and a testing set.
Figure 3 illustrates the description of k-NN-based genetic algorithm in accordance with an embodiment of the present disclosure. The first step is initializing population which is accomplished by generating Chromosome length by random binary digits based on maximum dataset length. Usually, chromosomes are denoted in binary values as strings of Os and Is. The second step includes generating children for the initial population in this step the genetic algorithm automatically picks the top two best chromosomes by using the elite-count size of two. Whereas, elite-count is less than or equal to the size of the population. In the third step the Genetic algorithm performs elitism, mutation and crossover on new population. In the fourth step fitness function of each chromosome is determined by using the k-NN algorithm. The fifth step includes selection mechanism in which Elite-count of size 2 is used because of its speed, simplicity and efficiency. The sixth step includes stopping criteria which is based on the number of iterations (100), when the generation reaches its predefined value; it stops and delivers the best solution in the last generation. Finally in the last step the classification accuracy of the new set of voxels is evaluated. These relevant voxels are given as the input for the GSW-LCDRC classifier for fMRI data classification.
Figure 4 illustrates a table of Performance comparison of the existing and proposed work in accordance with an embodiment of the present disclosure. The existing methodology of a fuzzy integral approach based on the ensemble neural networks for analyzing fMRI data for cognitive state classification across multiple subjects, achieved average 94.03% of classification accuracy, 51.75% of sensitivity and 94.87% of specificity with 80% training and 20% testing of data. Another existing methodology of a superior system for visualization, classification and dynamic learning of fMRI as spatiotemporal brain data achieved 87.5% of classification accuracy with 50% training and 50% testing of data. The developed system was based on a spatiotemporal data machine of evolving spiking neural networks (SNNs), which was exemplified by the NeuCube architecture. This research work was performed on a publicly available database (Star Plus dataset).
Another existing methodology SVM with confusion matrix for determining the warning signs of patient diseases achieved 67% of classification accuracy. The proposed disclosure
(GSW-LCDRC with k-NN-based genetic algorithm) was carried out on two fMRI datasets: StarPlus and Haxby dataset. However, the proposed disclosure achieved 92.14% of sensitivity, 91.09% of specificity and 90.71% of classification accuracy with 50% training and 50% testing of data. Additionally, the proposed disclosure achieved 99.90% of sensitivity, 99.74% of specificity, and 99.74% of classification accuracy with 80% training and 20% testing of medical data. The result of the proposed work was dramatically higher compared to the existing approaches.
Another existing solution of a new algorithm: hierarchical heterogeneous particle swarm optimization (HHPSO-SVMP) for multi-voxel pattern analysis achieved 83% of classification accuracy in multi-voxel pattern classification. Compared to this existing paper, the proposed methodology averagely achieved 88.95% of classification accuracy, which was almost 6% higher than the existing system.
The drawings and the forgoing description give examples of embodiments. Those skilled in the art will appreciate that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be split into multiple functional elements. Elements from one embodiment may be added to another embodiment. For example, orders of processes described herein may be changed and are not limited to the manner described herein. Moreover, the actions of any flow diagram need not be implemented in the order shown; nor do all of the acts necessarily need to be performed. Also, those acts that are not dependent on other acts may be performed in parallel with the other acts. The scope of embodiments is by no means limited by these specific examples. Numerous variations, whether explicitly given in the specification or not, such as differences in structure, dimension, and use of material, are possible. The scope of embodiments is at least as broad as given by the following claims.
Benefits, other advantages, and solutions to problems have been described above with regard to specific embodiments. However, the benefits, advantages, solutions to problems, and any component(s) that may cause any benefit, advantage, or solution to occur or become more pronounced are not to be construed as a critical, required, or essential feature or component of any or all the claims.
Claims (7)
1. A system for human cognitive states classification, the system comprises of:
a data collection module for collecting at least a dataset, wherein the dataset comprises of a Functional magnetic resonance imaging (fMRI) data arranged sequentially;
a feature selection module connected to the data collection module for identifying a relevant subset of fMRI data based on a particular criterion using a K-NN based genetic algorithm technique, wherein the genetic algorithm decreases redundancy within an input voxel and also determines the maximum relevance between an output and the input voxels, wherein the input voxel is calculated using a crossover and mutation operator; and
a classification module connected to the feature selection module for classifying the obtained fMRI data using a Linear Collaborative Discriminant Regression Classification (LCDRC) upon dividing the dataset into a training set and a testing set.
2. The system as claimed in claim 1, wherein the feature extraction module comprises of:
Generating an initial population of set as the subset of input voxels;
identifying a discriminative capability of each subset of the voxels using a fitness evaluator;
calculating fitness of each voxels using the K-NN algorithm technique; and
finding a shortest distance between the training and the testing set for classification, wherein the k-NN algorithm uses a Euclidean distance measure for identifying the nearest distance between the training and testing sets for reducing the classification error.
3. The system as claimed in claim 1, wherein crossover and mutation operators are used to find an optimal voxel by reducing the redundancy based on the fitness function.
4. The system as claimed 1, wherein initial population is accomplished using chromosome length is generated by random binary digits based on maximum dataset length, wherein the chromosomes are denoted in binary values as strings of Os and Is.
5. The system as claimed in claim 1, wherein the genetic algorithm automatically picks the top two best chromosomes by using an elite-count size of two, wherein the elite-count is less than or equal to the size of the population validating integrity of the files.
6. The system as claimed in claim 1, wherein a stopping criterion is applied during selection of the voxel, wherein the stopping criteria is determined by the number of iterations, when the generation reaches its predefined value; it stops and delivers the best solution in the last generation.
7. A method for human cognitive states classification, the method comprises of:
collecting at least a dataset using a data collection module, wherein the dataset comprises of a Functional magnetic resonance imaging (fMRI) data arranged sequentially;
identifying a relevant subset of fMRI data using a feature selection module connected to the data collection module based on a particular criterion using a K-NN based genetic algorithm technique, wherein the genetic algorithm decreases redundancy within an input voxel and also determines the maximum relevance between an output and the input voxels, wherein the input voxel is calculated using a crossover and mutation operator; and
classifying the obtained fMRI data using a Linear Collaborative Discriminant Regression Classification (LCDRC) of a classification module connected to the feature selection module upon dividing the dataset into a training set and a testing set.
Data collection module 102
Feature selection module 104
classification module 106
Figure 1
Figure 3
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| AU2021102810A AU2021102810A4 (en) | 2021-05-24 | 2021-05-24 | A system for human cognitive states classification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| AU2021102810A AU2021102810A4 (en) | 2021-05-24 | 2021-05-24 | A system for human cognitive states classification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| AU2021102810A4 true AU2021102810A4 (en) | 2021-12-02 |
Family
ID=78716453
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| AU2021102810A Ceased AU2021102810A4 (en) | 2021-05-24 | 2021-05-24 | A system for human cognitive states classification |
Country Status (1)
| Country | Link |
|---|---|
| AU (1) | AU2021102810A4 (en) |
-
2021
- 2021-05-24 AU AU2021102810A patent/AU2021102810A4/en not_active Ceased
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ayesha et al. | Overview and comparative study of dimensionality reduction techniques for high dimensional data | |
| Tanveer et al. | Comprehensive review on twin support vector machines | |
| Ray et al. | Various dimension reduction techniques for high dimensional data analysis: a review | |
| Tharwat et al. | Linear discriminant analysis: A detailed tutorial | |
| Ding et al. | A review on multi-class TWSVM | |
| Bai et al. | Time series classification based on multi-feature dictionary representation and ensemble learning | |
| Zhao et al. | Regularized Gaussian mixture model for high-dimensional clustering | |
| Huang et al. | Multivariate time series early classification using multi-domain deep neural network | |
| Zadeh et al. | High efficient system for automatic classification of the electrocardiogram beats | |
| Zhang et al. | Supervised feature selection algorithm via discriminative ridge regression | |
| Tavakoli | Seq2image: Sequence analysis using visualization and deep convolutional neural network | |
| CN114399634B (en) | Three-dimensional image classification method, system, equipment and medium based on weak supervision learning | |
| Arowolo et al. | A hybrid dimensionality reduction model for classification of microarray dataset | |
| Terol et al. | A machine learning approach to reduce dimensional space in large datasets | |
| Kumar et al. | An Analysis of Ensemble Machine Learning Algorithms for Breast Cancer Detection: Performance and Generalization | |
| Liang et al. | Multi-Scale Attention-Based Deep Neural Network for Brain Disease Diagnosis. | |
| Li et al. | Multi-label feature selection with high-sparse personalized and low-redundancy shared common features | |
| AU2021102810A4 (en) | A system for human cognitive states classification | |
| Androulakis et al. | Mathematical programming for data mining | |
| adnan Abdulrahman et al. | Using K-nearest neighbors and support vector machine classifiers in personal identification based on EEG signals | |
| Faiz et al. | Thyroid Disease Detection Using Machine Learning | |
| Iliyasu et al. | Evidence Accumulation Clustering with Possibilitic Fuzzy C-Means base clustering approach to disease diagnosis | |
| Jing et al. | GoogLeNet-like Model for Pedestrian Attribute Detection in Surveillance Environment | |
| Liu et al. | DSDCLNet: Dual-stream encoder and dual-level contrastive learning network for supervised multivariate time series classification | |
| Singla et al. | Smart Healthcare System for Disease Prediction using Hybrid Machine Learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| DA3 | Amendments made section 104 |
Free format text: THE NATURE OF THE AMENDMENT IS: AMEND THE NAME OF THE INVENTOR TO READ GUPTA, KAPIL AND CHATUR, PRASHANT |
|
| FGI | Letters patent sealed or granted (innovation patent) | ||
| MK22 | Patent ceased section 143a(d), or expired - non payment of renewal fee or expiry |