Interval feature extraction for classification of event-related potentials (ERP) in EEG data analysis

In the present work, a new method for the classification of Event Related Potentials (ERPs) is proposed. The proposed method consists of two modules: the feature extraction module and the classification module. The feature extraction module comprises the implementation of the Multivariate Autoregressive model in conjunction with the Simulated Annealing technique, for the selection of optimum features from ERPs. The classification module is implemented with a single three-layer neural network, trained with the back-propagation algorithm and classifies the data into two classes: patients and control subjects. The method, in the form of a Decision Support System (DSS), has been thoroughly tested to a number of patient data (OCD, FES, depressives and drug users), resulting successful classification up to 100%.

Download Full-text

Extracting Spatio-temporal Feature for Classification of Event-related Potentials*

PROGRESS IN BIOCHEMISTRY AND BIOPHYSICS ◽

10.3724/sp.j.1206.2011.00123 ◽

2011 ◽

Vol 38 (9) ◽

pp. 866-871 ◽

Cited By ~ 1

Author(s):

Zhi-Hua HUANG ◽

Ming-Hong LI ◽

Yuan-Ye MA ◽

Chang-Le ZHOU

Keyword(s):

Event Related Potentials ◽

Related Potentials ◽

Spatio Temporal ◽

Temporal Feature

Download Full-text

Eliciting and Recording Event Related Potentials (ERPs) in Behaviourally Unresponsive Populations: A Retrospective Commentary on Critical Factors

Brain Sciences ◽

10.3390/brainsci11070835 ◽

2021 ◽

Vol 11 (7) ◽

pp. 835

Author(s):

Alexander Rokos ◽

Richard Mah ◽

Rober Boshra ◽

Amabilis Harrison ◽

Tsee Leng Choy ◽

...

Keyword(s):

Event Related Potentials ◽

Stimulus Presentation ◽

Event Related Potential ◽

Critical Factors ◽

Healthy Elderly ◽

States Of Consciousness ◽

Long Latency ◽

Eeg Data ◽

Related Potentials ◽

The Relationship

A consistent limitation when designing event-related potential paradigms and interpreting results is a lack of consideration of the multivariate factors that affect their elicitation and detection in behaviorally unresponsive individuals. This paper provides a retrospective commentary on three factors that influence the presence and morphology of long-latency event-related potentials—the P3b and N400. We analyze event-related potentials derived from electroencephalographic (EEG) data collected from small groups of healthy youth and healthy elderly to illustrate the effect of paradigm strength and subject age; we analyze ERPs collected from an individual with severe traumatic brain injury to illustrate the effect of stimulus presentation speed. Based on these critical factors, we support that: (1) the strongest paradigms should be used to elicit event-related potentials in unresponsive populations; (2) interpretation of event-related potential results should account for participant age; and (3) speed of stimulus presentation should be slower in unresponsive individuals. The application of these practices when eliciting and recording event-related potentials in unresponsive individuals will help to minimize result interpretation ambiguity, increase confidence in conclusions, and advance the understanding of the relationship between long-latency event-related potentials and states of consciousness.

Download Full-text