Cross-spectral analysis of physiological tremor and muscle activity

Abstract

We investigate the relationship between the extensor electromyogram (EMG) and tremor times series in physiological hand tremor by cross-spectral analysis. Special attention is directed to the phase spectrum and the effects of observational noise. We calculate the theoretical phase spectrum for a second-order linear stochastic process and compare the results to measured tremor data recorded from subjects who did not show a synchronized EMG activity in the corresponding extensor muscle. The results show that physiological tremor is well described by the proposed model and that the measured EMG represents a Newtonian force by which the muscle acts on the hand.

Multiple oscillators are causing parkinsonian and essential tremor

“…The tools used for the estimation of these parameters and their confidence intervals are described in detail in other articles. 15,16,27,28 The raw EMG data were screened off-line. Sets of data containing movement artefacts as revealed by lowerfrequency baseline shifts were excluded from further analysis.…”

“…It has been shown on theoretical grounds that the coherence function is a more powerful tool with respect to the detection of a common origin of two signals. 15,16,27 This study presents the first data-based evidence that mere comparisons of tremor frequencies can sometimes underestimate and will generally overestimate the coupling between two recording sites.…”

“…Primary orthostatic tremor is the only form of pathologic tremor that is known to perfectly fulfill this criterion. All muscles taking part in the high-frequency tremor (14)(15)(16)(17)(18) are activated in a highly coherent manner regardless of the distance between the two muscles analyzed or whether they belong to different limbs. [31][32][33][34] Thus, orthostatic tremor originates, indeed, from a single central oscillator or oscillatory network.…”

“…[14][15][16] It is a means to detect a common input for the generation of two biologic signals and is therefore relevant for the study of relationships between the activity of two trembling muscles. In a previous study we used this method to compare the oscillations only in forearm muscles on both sides and found no side-to-side correlations.…”

The pathophysiology of parkinsonian tremor: a review

“…This has not been questioned until recently when new mathematical techniques have been applied to study these relations. The appropriate method to analyze such relations is coherence analysis [145,146,156] because it can separate between a condition when two muscles are only trembling at the same frequency or if they really share common inputs indicating a Fig. 1 Cross spectral analysis of three muscle combinations in a PD patient.…”