Spike Sorting via Multi Cluster Feature Selection

2016 ◽  
Author(s):  
Azadeh Dinparastdjadid ◽  
Ehsan T. Esfahani
Keyword(s):  
Feature Selection ◽  
Optimization Problem ◽  
Brain Computer Interface ◽  
Original Data ◽  
Computer Interface ◽  
Spike Sorting ◽  
Learning Problem ◽  
Data Set ◽  
Cognitive Sciences ◽  
Cluster Feature

Having the ability to study the activity of single neurons will facilitate studies in many areas including cognitive sciences and brain computer interface applications. Due to the fact that every neuron has it’s own unique spike waveform, by applying spike-sorting methods, one can separate neurons based on their associated spike. Spike sorting is an unsupervised learning problem in the realm of data mining and machine learning. In this study, a new method that will improve the accuracy of spike sorting in comparison to existing methods has been introduced. This method, which is named Multi Cluster Feature Selection (MCFS), will designate a reduced number of features from the original data set that will best differentiate the existing clusters through solving a Lasso optimization problem. MCFS, was also applied to data obtained from multi-channel recordings on a rat’s brain. With MCFS, each channel was studied and neurons in each channel were sorted with an improved rate in comparison to conventional methods such as PCA.

scholarly journals A Novel Fixed Low-Rank Constrained EEG Spatial Filter Estimation with Application to Movie-Induced Emotion Recognition

2016 ◽  
Vol 2016 ◽  
pp. 1-12 ◽  
Author(s):  
Ken Yano ◽  
Takayuki Suyama
Keyword(s):  
Feature Extraction ◽  
Feature Selection ◽  
Optimization Problem ◽  
Brain Computer Interface ◽  
Spatial Filter ◽  
Low Rank ◽  
Computer Interface ◽  
Emotional States ◽  
Loss Minimization ◽  
Spatial Filters

This paper proposes a novel fixed low-rank spatial filter estimation for brain computer interface (BCI) systems with an application that recognizes emotions elicited by movies. The proposed approach unifies such tasks as feature extraction, feature selection, and classification, which are often independently tackled in a “bottom-up” manner, under a regularized loss minimization problem. The loss function is explicitly derived from the conventional BCI approach and solves its minimization by optimization with a nonconvex fixed low-rank constraint. For evaluation, an experiment was conducted to induce emotions by movies for dozens of young adult subjects and estimated the emotional states using the proposed method. The advantage of the proposed method is that it combines feature selection, feature extraction, and classification into a monolithic optimization problem with a fixed low-rank regularization, which implicitly estimates optimal spatial filters. The proposed method shows competitive performance against the best CSP-based alternatives.

Feature selection for classification in Steady state visually evoked potentials (SSVEP)-based brain-computer interfaces with genetic algorithm

2020 ◽  
Vol 16 (2) ◽  
Author(s):  
Stanisław Karkosz ◽  
Marcin Jukiewicz
Keyword(s):  
Genetic Algorithm ◽  
Feature Selection ◽  
Steady State ◽  
Brain Computer Interface ◽  
Morphological Features ◽  
Computer Interface ◽  
Visually Evoked Potentials ◽  
Computer Interfaces ◽  
Subject Variability ◽  
Selection For

AbstractObjectivesOptimization of Brain-Computer Interface by detecting the minimal number of morphological features of signal that maximize accuracy.MethodsSystem of signal processing and morphological features extractor was designed, then the genetic algorithm was used to select such characteristics that maximize the accuracy of the signal’s frequency recognition in offline Brain-Computer Interface (BCI).ResultsThe designed system provides higher accuracy results than a previously developed system that uses the same preprocessing methods, however, different results were achieved for various subjects.ConclusionsIt is possible to enhance the previously developed BCI by combining it with morphological features extraction, however, it’s performance is dependent on subject variability.

Bayesian Classifier for Sparsity-Promoting Feature Selection

2015 ◽  
Vol 29 (06) ◽  
pp. 1550022 ◽  
Author(s):  
Danlei Xu ◽  
Lan Du ◽  
Hongwei Liu ◽  
Penghui Wang
Keyword(s):  
Feature Selection ◽  
Synthetic Data ◽  
Original Data ◽  
Radar Data ◽  
Bayesian Classifier ◽  
Classification Model ◽  
Data Sets ◽  
Data Set ◽  
Classification Boundary ◽  
Nonlinear Mappings

A Bayesian classifier for sparsity-promoting feature selection is developed in this paper, where a set of nonlinear mappings for the original data is performed as a pre-processing step. The linear classification model with such mappings from the original input space to a nonlinear transformation space can not only construct the nonlinear classification boundary, but also realize the feature selection for the original data. A zero-mean Gaussian prior with Gamma precision and a finite approximation of Beta process prior are used to promote sparsity in the utilization of features and nonlinear mappings in our model, respectively. We derive the Variational Bayesian (VB) inference algorithm for the proposed linear classifier. Experimental results based on the synthetic data set, measured radar data set, high-dimensional gene expression data set, and several benchmark data sets demonstrate the aggressive and robust feature selection capability and comparable classification accuracy of our method comparing with some other existing classifiers.

Feature selection method based on Menger curvature and LDA theory for a P300 brain-computer interface

2021 ◽  
Author(s):  
ShuRui Li ◽  
Jing Jin ◽  
Ian Daly ◽  
Chang Liu ◽  
Andrzej Cichocki
Keyword(s):  
Feature Selection ◽  
Brain Computer Interface ◽  
Feature Selection Method ◽  
Event Related Potentials ◽  
Selection Method ◽  
Computer Interface ◽  
Feature Subset ◽  
Linear Discriminant ◽  
Related Potentials ◽  
Menger Curvature

Abstract Brain–computer interface (BCI) systems decode electroencephalogram signals to establish a channel for direct interaction between the human brain and the external world without the need for muscle or nerve control. The P300 speller, one of the most widely used BCI applications, presents a selection of characters to the user and performs character recognition by identifying P300 event-related potentials from the EEG. Such P300-based BCI systems can reach good levels of accuracy but are difficult to use in day-to-day life due to redundancy and noisy signal. A room for improvement should be considered. We propose a novel hybrid feature selection method for the P300-based BCI system to address the problem of feature redundancy, which combines the Menger curvature and linear discriminant analysis. First, selected strategies are applied separately to a given dataset to estimate the gain for application to each feature. Then, each generated value set is ranked in descending order and judged by a predefined criterion to be suitable in classification models. The intersection of the two approaches is then evaluated to identify an optimal feature subset. The proposed method is evaluated using three public datasets, i.e., BCI Competition III dataset II, BNCI Horizon dataset, and EPFL dataset. Experimental results indicate that compared with other typical feature selection and classification methods, our proposed method has better or comparable performance. Additionally, our proposed method can achieve the best classification accuracy after all epochs in three datasets. In summary, our proposed method provides a new way to enhance the performance of the P300-based BCI speller.

scholarly journals Feature Selection and Blind Source Separation in an EEG-Based Brain-Computer Interface

2005 ◽  
Vol 2005 (19) ◽  
Author(s):  
David A. Peterson ◽  
James N. Knight ◽  
Michael J. Kirby ◽  
Charles W. Anderson ◽  
Michael H. Thaut
Keyword(s):  
Feature Selection ◽  
Blind Source Separation ◽  
Brain Computer Interface ◽  
Source Separation ◽  
Computer Interface
Sign in / Sign up

Export Citation Format

Share Document