DCCA multi cross-correlation analysis applied on EEG signals to study motor activity (Real/Imaginary)

• Published in: Biomedical signal processing and control Vol. 103; p. 107419
• Main Authors: Ribeiro, Fernando Ferraz, Brito, Andréa de Almeida, Filho, Florêncio Mendes Oliveira, Cruz, Juan Alberto Leyva, Zebende, Gilney Figueira
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.05.2025
• Subjects: Brain; DCCA; DCCA multiple cross-correlation coefficient; DFA; EEG signals; Brain; DCCA; EEG signals; DCCA multiple cross-correlation coefficient; DFA
• ISSN: 1746-8094
• DOI: 10.1016/j.bspc.2024.107419

We applied the DCCA multiple cross-correlation coefficient to analyze time series of EEG experiment, where 109 subjects performed four tasks involving real and imaginary motor activities. In this case, four specific channels were selected on the scalp: two on the frontal and two on parietal region. As a result, the DCCA multiple cross-correlation coefficient identified that there is a single signature for each subject. Globally, there is no significant difference between the real and the imaginary task. The frontal channels had greater multiple cross-correlation values than the parietal ones, especially for time scales around 0.42s, and with smaller standard deviations. According to our results, the proposal to study multiple time series at the same time by using the DCCA multiple cross-correlation coefficient is feasible and robust for EEG analysis. Finally, the coefficient was applied in a substantial number of subjects, tasks, and experiments producing high-quality results (figures, movies, and tables), and this study will probably kick off of a new approach to analyzing multiple cross-correlations in EEG signals. •We propose to study the EEG bio-signals in the control of real/imaginary tasks of locomotion or movement of the human being.•Four EEG channels was analyzed in the brain with DMCx2 coefficient.•The multiple coefficient identifies that a unique signature for each subject appears.•As a final result, we obtained a global and a individual statistics.