Time-frequency analysis of spontaneous pupillary oscillation signals using the Hilbert-Huang transform. (September 2016)

• Record Type: Journal Article
• Authors: Villalobos-Castaldi, Fabiola M.; Ruiz-Pinales, José; Valverde, Nicolás C. Kemper; Flores, Mercedes
• Is Part Of: Biomedical signal processing and control. Volume 30(2016)
• Page Start: 106
• Page End: 116
• Publication Date: 2016-09
• Subjects: Time-frequency analysis -- Hilbert-Huang transform -- Spontaneous pupillary oscillation -- Non-traditional time-series characterization scheme; 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.2016.06.002
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: Time-Frequency characterization of spontaneous pupillary fluctuatio. A non-traditional alternative for analyzing non-stationary signals. Introduction to the Hilbert-Huang Transform HHT in pupillometry. Amplitude-time-frequency representation of spontaneous pupillary oscillation. Results indicate that HHT is a reliable and objective tool in the detection and quantitatively description of spontaneous pupillary oscillation. Abstract: In this work a new application of the Hilbert-Huang transform (HHT) is proposed. Recordings of pupillograms were analyzed through the non-traditional HHT, in order to investigate patterns in the time-frequency parameters of the spontaneous pupillary oscillation (SPO) signals. Unlike the traditional Fourier transform, which is only useful for the analysis of linear stationary signals, the HHT was designed for the analysis of non-linear and non-stationary signals. Therefore, the HHT is a more suitable tool to study SPO signals which are fundamentally non-stationary. The intrinsic properties of the spontaneous pupillary oscillation signals were characterized under the HHT scheme. The obtained results showed that SPO waves exhibit local and intermittent variations through the time span. From the numerical parameters that were obtained, we could observe that there is a wide inter-subject variation in the contribution of the intrinsic time-frequency parameters from each yielding mode to the total signal content.