Diffuse to fuse EEG spectra – Intrinsic geometry of sleep dynamics for classification

Gi Ren Liu, Yu Lun Lo, John Malik, Yuan Chung Sheu, Hau Tieng Wu

Research output: Contribution to journalArticle

Abstract

We propose a novel algorithm for sleep dynamics visualization and automatic annotation by applying diffusion geometry based sensor fusion algorithm to fuse spectral information from two electroencephalograms (EEG). The diffusion geometry approach helps organize the nonlinear dynamical structure hidden in the EEG signal. The visualization is achieved by the nonlinear dimension reduction capability of the chosen diffusion geometry algorithms. For the automatic annotation purpose, the support vector machine is trained to predict the sleep stage. The prediction performance is validated on a publicly available benchmark database, Physionet Sleep-EDF [extended] SC* (SC = Sleep Cassette) and ST* (ST = Sleep Telemetry), with the leave-one-subject-out cross validation. When we have a single EEG channel (Fpz-Cz), the overall accuracy, macro F1 and Cohen's kappa achieve 82.72%, 75.91% and 76.1% respectively in Sleep-EDF SC* and 78.63%, 73.58% and 69.48% in Sleep-EDF ST*. This performance is compatible with the state-of-the-art results. When we have two EEG channels (Fpz-Cz and Pz-Oz), the overall accuracy, macro F1 and Cohen's kappa achieve 84.44%, 78.25% and 78.36% respectively in Sleep-EDF SC* and 79.05%, 74.73% and 70.31% in Sleep-EDF ST*. The results suggest the potential of the proposed algorithm in practical applications.

Original languageEnglish
Article number101576
JournalBiomedical Signal Processing and Control
Volume55
DOIs
Publication statusPublished - 2020 Jan

Fingerprint

Electric fuses
Electroencephalography
Sleep
Geometry
Telemetry
Telemetering
Macros
Visualization
Benchmarking
Sleep Stages
Support vector machines
Fusion reactions
Databases

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Health Informatics

Cite this

@article{4c9eeb0ef4094f839030c5a66121c96d,
title = "Diffuse to fuse EEG spectra – Intrinsic geometry of sleep dynamics for classification",
abstract = "We propose a novel algorithm for sleep dynamics visualization and automatic annotation by applying diffusion geometry based sensor fusion algorithm to fuse spectral information from two electroencephalograms (EEG). The diffusion geometry approach helps organize the nonlinear dynamical structure hidden in the EEG signal. The visualization is achieved by the nonlinear dimension reduction capability of the chosen diffusion geometry algorithms. For the automatic annotation purpose, the support vector machine is trained to predict the sleep stage. The prediction performance is validated on a publicly available benchmark database, Physionet Sleep-EDF [extended] SC* (SC = Sleep Cassette) and ST* (ST = Sleep Telemetry), with the leave-one-subject-out cross validation. When we have a single EEG channel (Fpz-Cz), the overall accuracy, macro F1 and Cohen's kappa achieve 82.72{\%}, 75.91{\%} and 76.1{\%} respectively in Sleep-EDF SC* and 78.63{\%}, 73.58{\%} and 69.48{\%} in Sleep-EDF ST*. This performance is compatible with the state-of-the-art results. When we have two EEG channels (Fpz-Cz and Pz-Oz), the overall accuracy, macro F1 and Cohen's kappa achieve 84.44{\%}, 78.25{\%} and 78.36{\%} respectively in Sleep-EDF SC* and 79.05{\%}, 74.73{\%} and 70.31{\%} in Sleep-EDF ST*. The results suggest the potential of the proposed algorithm in practical applications.",
author = "Liu, {Gi Ren} and Lo, {Yu Lun} and John Malik and Sheu, {Yuan Chung} and Wu, {Hau Tieng}",
year = "2020",
month = "1",
doi = "10.1016/j.bspc.2019.101576",
language = "English",
volume = "55",
journal = "Biomedical Signal Processing and Control",
issn = "1746-8094",
publisher = "Elsevier BV",

}

Diffuse to fuse EEG spectra – Intrinsic geometry of sleep dynamics for classification. / Liu, Gi Ren; Lo, Yu Lun; Malik, John; Sheu, Yuan Chung; Wu, Hau Tieng.

In: Biomedical Signal Processing and Control, Vol. 55, 101576, 01.2020.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Diffuse to fuse EEG spectra – Intrinsic geometry of sleep dynamics for classification

AU - Liu, Gi Ren

AU - Lo, Yu Lun

AU - Malik, John

AU - Sheu, Yuan Chung

AU - Wu, Hau Tieng

PY - 2020/1

Y1 - 2020/1

N2 - We propose a novel algorithm for sleep dynamics visualization and automatic annotation by applying diffusion geometry based sensor fusion algorithm to fuse spectral information from two electroencephalograms (EEG). The diffusion geometry approach helps organize the nonlinear dynamical structure hidden in the EEG signal. The visualization is achieved by the nonlinear dimension reduction capability of the chosen diffusion geometry algorithms. For the automatic annotation purpose, the support vector machine is trained to predict the sleep stage. The prediction performance is validated on a publicly available benchmark database, Physionet Sleep-EDF [extended] SC* (SC = Sleep Cassette) and ST* (ST = Sleep Telemetry), with the leave-one-subject-out cross validation. When we have a single EEG channel (Fpz-Cz), the overall accuracy, macro F1 and Cohen's kappa achieve 82.72%, 75.91% and 76.1% respectively in Sleep-EDF SC* and 78.63%, 73.58% and 69.48% in Sleep-EDF ST*. This performance is compatible with the state-of-the-art results. When we have two EEG channels (Fpz-Cz and Pz-Oz), the overall accuracy, macro F1 and Cohen's kappa achieve 84.44%, 78.25% and 78.36% respectively in Sleep-EDF SC* and 79.05%, 74.73% and 70.31% in Sleep-EDF ST*. The results suggest the potential of the proposed algorithm in practical applications.

AB - We propose a novel algorithm for sleep dynamics visualization and automatic annotation by applying diffusion geometry based sensor fusion algorithm to fuse spectral information from two electroencephalograms (EEG). The diffusion geometry approach helps organize the nonlinear dynamical structure hidden in the EEG signal. The visualization is achieved by the nonlinear dimension reduction capability of the chosen diffusion geometry algorithms. For the automatic annotation purpose, the support vector machine is trained to predict the sleep stage. The prediction performance is validated on a publicly available benchmark database, Physionet Sleep-EDF [extended] SC* (SC = Sleep Cassette) and ST* (ST = Sleep Telemetry), with the leave-one-subject-out cross validation. When we have a single EEG channel (Fpz-Cz), the overall accuracy, macro F1 and Cohen's kappa achieve 82.72%, 75.91% and 76.1% respectively in Sleep-EDF SC* and 78.63%, 73.58% and 69.48% in Sleep-EDF ST*. This performance is compatible with the state-of-the-art results. When we have two EEG channels (Fpz-Cz and Pz-Oz), the overall accuracy, macro F1 and Cohen's kappa achieve 84.44%, 78.25% and 78.36% respectively in Sleep-EDF SC* and 79.05%, 74.73% and 70.31% in Sleep-EDF ST*. The results suggest the potential of the proposed algorithm in practical applications.

UR - http://www.scopus.com/inward/record.url?scp=85073684433&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073684433&partnerID=8YFLogxK

U2 - 10.1016/j.bspc.2019.101576

DO - 10.1016/j.bspc.2019.101576

M3 - Article

AN - SCOPUS:85073684433

VL - 55

JO - Biomedical Signal Processing and Control

JF - Biomedical Signal Processing and Control

SN - 1746-8094

M1 - 101576

ER -