A novel analytical method to measure intra-individual variability of steady-state evoked potentials; new insights into attention deficit. (January 2022)

• Record Type: Journal Article
• Authors: Norouzpour, Amir; Klein, Stanley A.
• Is Part Of: Biomedical signal processing and control. Volume 71(2022)Part A
• Publication Date: 2022-01
• Subjects: Steady-State Evoked potential -- Attention -- Covariance -- Eigenvector -- Attention deficit hyperactive disorder -- Ellipse; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2021.103109
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: Variation in amplitude and phase of frequency component taken from SSEP is measured. We fitted the Fourier estimates using an ellipse instead of a circle. Covariance to eigenvectors conversion yields the ellipse parameters. The ellipse Length-Width difference reveals meaningful fluctuations in brain response. The Angle of the ellipse estimates the correlation between the amplitude and phase. Abstract: Background: The intra-individual variability of steady-state evoked potential (SS-EP) recordings reveals unique individual behavioral capabilities in neuroscience experiments and individual attention deficit in clinical studies. In this paper, we show how to measure the intra-individual variability in both amplitude and phase of frequency component. Methods: To estimate the variability, multiple Fourier measurements at a frequency component using a non-overlapping moving time window are extracted from SS-EP data and represented in the two-dimensional complex plane. In contrast to the common approach to fit the Fourier estimates with a circle based on the assumption that the brain response is stable over time, our approach based on the reasonable assumption of fluctuating attention is to fit the samples using an ellipse with three new parameters: Length, Width, and Angle of the ellipse. We believe that attention fluctuations can produce a significant Length-Width difference (LW-Diff). Results: We applied our method to simulations to demonstrate validity. If the LW-Diff is significant using standard error of standard deviation, the cluster of Fourier estimates is non-circular in the complex plane. In such cases, we assume the maximum variation corresponds to a meaningful fluctuation in brain response presumably as a function of attention. Conclusion: Our method enables us, for the first time, to estimate individual attention fluctuation using the Length-to-Width Ratio and to test whether Fourier measurements distribute circularly in the complex plane. The method is useful for distinguishing individuals with different behavioral capabilities, differentiating cases with inconsistent control of attention from healthy subjects, and monitoring therapeutic interventions.