scholarly journals Mental Arithmetic Task Recognition Using Effective Connectivity and Hierarchical Feature Selection from EEG Signals

2021 ◽  
Author(s):  
Ahmad Shalbaf ◽  
◽  
Arash Maghsoudi ◽  
Keyword(s):  
Feature Selection ◽  
Transfer Function ◽  
Mental Arithmetic ◽  
Brain Connectivity ◽  
Effective Connectivity ◽  
Concave Minimization ◽  
Autism Spectrum ◽  
Feature Subset Selection ◽  
Eeg Signal ◽  
Directed Transfer Function

Introduction: Mental arithmetic analysis based on Electroencephalogram (EEG) signal for monitoring the state of the user’s brain functioning can be helpful for understanding some psychological disorders such as attention deficit hyperactivity disorder, autism spectrum disorder, or dyscalculia where the difficulty in learning or understanding the arithmetic exists. Most mental arithmetic recognition systems rely on features of a single channel of EEG; however, the relationships among EEG channels in the form of effective brain connectivity analysis can contain valuable information. The aim of this paper is to identify a set of discriminative effective brain connectivity features from EEG signal and develop a hierarchical feature selection structure for classification of mental arithmetic and baseline tasks effectively. Methods: We estimated effective connectivity using Directed Transfer Function (DTF), direct Directed Transfer Function (dDTF) and Generalized Partial Directed Coherence (GPDC) methods. These measures determine the causal relation between different brain areas. To select most significant effective connectivity features, a hierarchical feature subset selection method is used. First Kruskal–Wallis test was performed and consequently, five feature selection algorithms namely Support Vector Machine ( SVM ) method based on Recursive Feature Elimination, Fisher score, mutual information, minimum Redundancy Maximum Relevance and concave minimization and SVM are used to select the best discriminative features. Finally, SVM method was used for classification. Results: Results show that the best EEG classification performance in 29 participants and 60 trials is obtained using GPDC and feature selection via concave minimization method in Beta2 (15−22Hz) frequency band with 89% accuracy. Conclusion: This new hierarchical automated system could be useful for discrimination of mental arithmetic and baseline tasks from EEG signal effectively.

scholarly journals Classification of Right/Left Hand Motor Imagery by Effective Connectivity Based on Transfer Entropy in EEG Signal

2021 ◽  
pp. 1-22
Author(s):  
Erfan Rezaei ◽  
◽  
Ahmad Shalbaf ◽  
Keyword(s):  
Feature Selection ◽  
Motor Imagery ◽  
Effective Connectivity ◽  
Feature Subset Selection ◽  
Transfer Entropy ◽  
Support Vector ◽  
Feature Subset ◽  
Eeg Signal ◽  
Left Hand ◽  
Right And Left Hand

The right and left hand Motor Imagery (MI) analysis based on the electroencephalogram (EEG) signal can directly link the central nervous system to a computer or a device. This study aims to identify a set of robust and nonlinear effective brain connectivity features quantified by transfer entropy (TE) to characterize the relationship between brain regions from EEG signals and create a hierarchical feature selection and classification for discrimination of right and left hand MI task. TE is calculated among EEG channels as the distinctive, effective connectivity features. TE is a model-free method that can measure nonlinear effective connectivity and analyze multivariate dependent directed information flow among neural EEG channels. Then four feature subset selection methods namely Relief-F, Fisher, Laplacian and local learning based clustering (LLCFS) algorithms are used to choose the most significant effective connectivity features and reduce redundant information. Finally, support vector machine (SVM) and LDA methods are used for classification. Results show that the best performance in 29 healthy subjects and 60 trials is achieved using the TE method via Relief-F algorithm as feature selection and SVM classification with 91.02% accuracy. Consequently, TE index and a hierarchical feature selection and classification could be useful for discrimination of right and left hand MI task from multichannel EEG signal.

scholarly journals Structurally constrained effective brain connectivity

2021 ◽  
Author(s):  
Alessandro Crimi
Keyword(s):  
Community Detection ◽  
Autoregressive Model ◽  
Brain Connectivity ◽  
Effective Connectivity ◽  
Structural Connectivity ◽  
Ground Truth ◽  
Autism Spectrum ◽  
Case Control ◽  
Functional Dependencies ◽  
The Brain

The relationship between structure and function is of interest in many research fields involving the study of complex biological processes. In neuroscience in particular, the fusion of structural and functional data can help to understand the underlying principles of the operational networks in the brain. To address this issue, this paper proposes a constrained autoregressive model leading to a representation of effective connectivity that can be used to better understand how the structure modulates the function. Or simply, it can be used to find novel biomarkers characterizing groups of subjects. In practice, an initial structural connectivity representation is re-weighted to explain the functional co-activations. This is obtained by minimizing the reconstruction error of an autoregressive model constrained by the structural connectivity prior. The model has been designed to also include indirect connections, allowing to split direct and indirect components in the functional connectivity, and it can be used with raw and deconvoluted BOLD signal.The derived representation of dependencies was compared to the well known dynamic causal model, giving results closer to known ground-truth. Further evaluation of the proposed effective network was performed on two typical tasks. In a first experiment the direct functional dependencies were tested on a community detection problem, where the brain was partitioned using the effective networks across multiple subjects. In a second experiment the model was validated in a case-control task, which aimed at differentiating healthy subjects from individuals with autism spectrum disorder. Results showed that using effective connectivity leads to clusters better describing the functional interactions in the community detection task, while maintaining the original structural organization, and obtaining a better discrimination in the case-control classification task.

scholarly journals Validation of a new approach for distinguishing anesthetized from awake state in patients using Directed Transfer Function applied to raw EEG

2020 ◽  
Author(s):  
Bjørn E. Juel ◽  
Luis Romundstad ◽  
Johan F. Storm ◽  
Pål G. Larsson
Keyword(s):  
Transfer Function ◽  
General Anesthesia ◽  
Molecular Mechanisms ◽  
Brain Connectivity ◽  
High Sensitivity ◽  
Classification Algorithm ◽  
The State ◽  
High Temporal Resolution ◽  
Directed Transfer Function ◽  
Significant Difference

AbstractAimIn a previous study, we found that the state of wakefulness in patients undergoing general anesthesia with propofol can effectively be monitored with high temporal resolution using an automatic measure of connectivity based on the Directed Transfer Function (DTF) calculated from short segments of electroencephalography (EEG) time-series. The study described here was designed to test whether the same measure can be used to monitor the state of the patients also during sevoflurane anesthesia.MethodsTwenty-five channel EEG recordings were collected from 8 patients undergoing surgical anesthesia with sevoflurane. The EEG data were segmented into one second epochs and labeled as awake or anesthetized in accordance with the clinician’s judgement, and the sensor space directed connectivity was quantified for every epoch using the DTF. The resulting DTF derived connectivity parameters were compared to corresponding parameters from our previous study using permutation statistics. A data driven classification algorithm was then employed to objectively classify the individual 1-second epochs as coming from awake or anesthetized state, using a leave-one-out cross-validation approach. The classifications were made for every epoch using the median DTF parameters across the five preceding 1-second EEG epochs.ResultsThe DTF derived connectivity parameters showed a significant difference between the awake and sevoflurane-induced general anesthesia at the group level (p<0.05). In contrast, the DTF parameters were not significantly different when comparing sevoflurane and propofol data neither in the awake nor in anesthetized state (p>0.05 for both comparisons). The classification algorithm reached a maximum accuracy of 96.8% (SE=0.63%). Optimizing the algorithm for simultaneously having high sensitivity and specificity in classification reduced the accuracy to 95.1% (SE=0.96%), with sensitivity of 98.4% (SE=0.80%) and specificity of 94.8% (SE=0.10%).ConclusionThese findings indicate that the DTF changes in a similar manner when humans undergo general anesthesia caused by two distinct anesthetic agents with different molecular mechanisms of action. This seems to support the idea that brain connectivity is related to the level of consciousness in humans, although further studies are needed to clarify whether our results may be contaminated by confounding factors.

Analysis of Speech Imagery using Functional and Effective EEG based Brain Connectivity Parameters

2015 ◽  
Vol 9 (4) ◽  
pp. 33-48 ◽  
Author(s):  
Sandhya Chengaiyan ◽  
Kavitha Anandhan
Keyword(s):  
Brain Connectivity ◽  
Effective Connectivity ◽  
Imagery Task ◽  
Left Frontal Lobe ◽  
Directed Transfer Function ◽  
Invasive Approach ◽  
Non Invasive ◽  
Wernicke’S Area ◽  
The Given ◽  
The Brain

Speech imagery is a form of mental imagery which refers to the activity of talking to oneself in silence. In this paper, EEG coherence, a functional connectivity parameter is calculated to analyze the concurrence of the different regions of the brain and Effective connectivity parameters such as Partial Directed Coherence (PDC), Directed Transfer Function (DTF) and Information theory based parameter Transfer Entropy (TE) are estimated to find the direction and strength of the connectivity patterns of the given speech imagery task. It has been observed from the results that by using functional and effective connectivity parameters the left frontal lobe electrodes was found to be high during speech production and left temporal lobe electrodes was found to be high while imagining the word silently in the brain due to the proximity of the electrodes to the Broca's and Wernicke's area respectively. The results suggest that the proposed methodology is a promising non-invasive approach to study directional connectivity in the brain between mutually interconnected neural populations.

scholarly journals Structurally Constrained Effective Brain Connectivity

2018 ◽  
Author(s):  
Alessandro Crimi ◽  
Luca Dodero ◽  
Fabio Sambataro ◽  
Vittorio Murino ◽  
Diego Sona
Keyword(s):  
Autoregressive Model ◽  
Brain Research ◽  
Brain Connectivity ◽  
Effective Connectivity ◽  
Structural Connectivity ◽  
Young Patients ◽  
Autism Spectrum ◽  
Case Control ◽  
Brain Targeting ◽  
The Brain

How function arises from structure is of interest in many fields from proteomics to neuroscience. In particular, among the brain research community the fusion of structure and function data can shed new lights on underlying operational network principles in the brain. Targeting this issue, the manuscript proposes a constrained autoregressive model generating “effective” connectivity given structural and functional information. In practice, an initial structural connectivity representation is altered according to functional data, by minimizing the reconstruction error of an autoregressive model constrained by the structural prior. The proposed model has been tested in a community detection framework, where the brain is partitioned using the effective networks across multiple subjects. The model is further validated in a case-control experiment, which aims at differentiating healthy subjects from young patients affected by autism spectrum disorder. Results showed that using effective connectivity resulted in clusters that better describe the functional interactions between different regions while maintaining the structural organization, and a better discrimination in the case-control classification task.

Sign in / Sign up

Export Citation Format

Share Document