نهمین کنگره نقشه برداری مغز

نمایش مقاله

Title:

EEGLAB-FDA: A Functional Data Analysis Framework for Advanced EEG Signal Processing

Subject:

Artificial Intelligence and Brain Mapping

Presenter Author:
Name:	Mohammad Fayaz

Author:
Name 1:	Mohammad Fayaz - Corresponding - Presenter
Affiliation 1:	Assistant Professor of Biostatistics, School of Medicine, Shahed University, Tehran, Iran. ORCID: 0000-0002-5643-9763

Name 2:
Affiliation 2:

Name 3:
Affiliation 3:

Name 4:
Affiliation 4:

Name 5:
Affiliation 5:

Name 6:
Affiliation 6:

Name 7:
Affiliation 7:

Name 8:
Affiliation 8:

Name 9:
Affiliation 9:

Name 10:
Affiliation 10:

Abstract

Original Research

introduction :

Electroencephalography (EEG) provides high-temporal-resolution measurements of brain electrical activity and is widely used in neuroscience, cognitive science, and clinical research. However, conventional EEG analysis often treats recordings as discrete observations, potentially overlooking the continuous functional structure of neural signals. Functional Data Analysis (FDA) provides a statistical framework for representing EEG recordings as smooth functions and enables the investigation of temporal patterns, functional variability, covariance structures, and associations between brain signals. Despite its potential, FDA methods are not routinely integrated into established EEG analysis environments, creating a methodological and computational gap for researchers who require accessible tools for functional analysis of EEG data.

Methods:

This project aims to develop and evaluate EEGLAB-FDA, an open-source MATLAB plug-in that extends the EEGLAB environment with a comprehensive set of Functional Data Analysis methods for EEG signals. The framework represents EEG recordings using basis-function expansions and provides tools for functional data preprocessing, functional principal component analysis (FPCA), functional canonical correlation analysis (fCCA), functional covariance and correlation analysis, spectral analysis, and functional bootstrap procedures. B-spline and Fourier basis representations are incorporated to model different temporal characteristics of EEG signals. The framework is designed to integrate established FDA methodologies, including approaches implemented through the fdaM and PACE MATLAB packages, within an EEG-oriented graphical workflow. The software will be evaluated using simulated and empirical EEG datasets, with emphasis on computational performance, reproducibility, usability, and consistency with established statistical implementations. Outputs will be assessed through graphical and numerical representations of functional trajectories, principal components, covariance structures, and frequency-domain characteristics.

Results:

The proposed framework is expected to provide an integrated computational environment for applying FDA methods directly to EEG recordings while preserving the established EEGLAB workflow. It will enable researchers to analyze EEG signals as continuous functional objects rather than relying exclusively on point-wise or manually selected temporal features. The framework is expected to facilitate the identification of dominant temporal patterns, functional sources of variation, relationships between EEG signal sets, and frequency-domain characteristics. Evaluation against simulated and empirical datasets will provide evidence regarding the numerical behavior, reproducibility, and practical applicability of the implemented methods. By combining FDA methodology with the widely used EEGLAB environment, EEGLAB-FDA is intended to reduce technical barriers to the application of functional statistical methods in EEG research.

Conclusion:

EEGLAB-FDA provides a methodological and computational framework for integrating Functional Data Analysis into EEG signal processing. By combining functional statistical modeling with an established EEG analysis environment, the framework can support more comprehensive characterization of the temporal structure and variability of neural signals. The resulting open-source software is expected to facilitate reproducible FDA-based EEG analysis and provide researchers with additional statistical tools for investigating complex brain dynamics.

Keywords:

EEGLAB-FDA; Functional Data Analysis; EEG; Functional Principal Component Analysis; EEG Signal Processing