loading page

Brain Evoked Response Qualification Using Multi-set Consensus Clustering: Toward Single-trial EEG Analysis
  • +4
  • Reza Mahini,
  • Guanghui Zhang,
  • Tiina Parviainen,
  • Rainer Düsing,
  • Asoke Nandi,
  • Fengyu Cong,
  • Timo Hämäläinen
Reza Mahini
University of Jyvaskyla Faculty of Mathematics and Science

Corresponding Author:remahini@jyu.fi

Author Profile
Guanghui Zhang
University of California Davis
Author Profile
Tiina Parviainen
University of Jyväskylä
Author Profile
Rainer Düsing
Osnabrück University
Author Profile
Asoke Nandi
Brunel University London
Author Profile
Fengyu Cong
Dalian University of Technology
Author Profile
Timo Hämäläinen
University of Jyvaskyla Faculty of Mathematics and Science
Author Profile

Abstract

Objective: Scalp electroencephalogram (EEG) provides a substantial amount of data about information processing in the human brain. In the context of conventional event-related potential (ERP analysis), it is typically assumed that individual trials share similar properties and stem from comparable neural sources, especially when employing group-level methods (including cluster analysis). However, those group analyses can miss important information about the relevant neural process due to a rough estimation of the brain activities of individual subjects while selecting a fixed time window for all the subjects. Method: We designed a multi-set consensus clustering method to examine cognitive processes at the single-trial level. The obtained clusters for the trials were processed via consensus clustering at the individual subject level. The proposed method effectively identified the time window of interest for each individual subject. Results: The proposed method was applied to real EEG data from the active visual oddball task experiment to qualify the P3 component. Our early findings disclosed that the estimated time windows for individual subjects can provide more precise ERP identification than considering a fixed time window for all subjects. Moreover, based on standardized measurement error and established bootstrap for single-trial EEG, our assessments revealed suitable stability for the calculated scores for the identified P3 component. Significance: The new method provides a more realistic and information-driven understanding of the single trials’ contribution towards identifying the ERP of interest in individual ERP potential data.