Empirical mode decomposition: A novel technique for the study of tremor time series

Eduardo Rocon de Lima*, Adriano O. Andrade, Jose L Pons, Peter Kyberd, Slawomir J. Nasuto

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

64 Scopus citations

Abstract

Tremor is a clinical feature characterized by oscillations of a part of the body. The detection and study of tremor is an important step in investigations seeking to explain underlying control strategies of the central nervous system under natural (or physiological) and pathological conditions. It is well established that tremorous activity is composed of deterministic and stochastic components. For this reason, the use of digital signal processing techniques (DSP) which take into account the nonlinearity and nonstationarity of such signals may bring new information into the signal analysis which is often obscured by traditional linear techniques (e.g. Fourier analysis). In this context, this paper introduces the application of the empirical mode decomposition (EMD) and Hilbert spectrum (HS), which are relatively new DSP techniques for the analysis of nonlinear and nonstationary time-series, for the study of tremor. Our results, obtained from the analysis of experimental signals collected from 31 patients with different neurological conditions, showed that the EMD could automatically decompose acquired signals into basic components, called intrinsic mode functions (IMFs), representing tremorous and voluntary activity. The identification of a physical meaning for IMFs in the context of tremor analysis suggests an alternative and new way of detecting tremorous activity. These results may be relevant for those applications requiring automatic detection of tremor. Furthermore, the energy of IMFs was visualized as a function of time and frequency by means of the HS. This analysis showed that the variation of energy of tremorous and voluntary activity could be distinguished and characterized on the HS. Such results may be relevant for those applications aiming to identify neurological disorders. In general, both the HS and EMD demonstrated to be very useful to perform objective analysis of any kind of tremor and can therefore be potentially used to perform functional assessment.

Original languageEnglish (US)
Pages (from-to)569-582
Number of pages14
JournalMedical and Biological Engineering and Computing
Volume44
Issue number7
DOIs
StatePublished - Jul 2006

Keywords

  • Empirical mode decomposition
  • Hilbert spectrum
  • Inertial sensors
  • Time-frequency analysis
  • Tremor
  • Tremor diagnostic
  • Tremor quantification

ASJC Scopus subject areas

  • Biomedical Engineering
  • Computer Science Applications

Fingerprint Dive into the research topics of 'Empirical mode decomposition: A novel technique for the study of tremor time series'. Together they form a unique fingerprint.

Cite this