Detection of tremor bursts by a running second order moment function and analysis using interburst histograms

Area of Science:

• Biomedical Engineering
• Signal Processing
• Neuroscience

Background:

• Conventional linear signal processing struggles with electromyographic (EMG) burst detection in pathological tremors due to noise.
• Non-linear analysis offers a potential solution for improved accuracy in clinical settings.

Purpose of the Study:

• To introduce and evaluate a non-linear analysis technique for detecting tremor bursts in EMG signals.
• To assess the utility of auto- and cross-interburst interval histograms (IBIH) in characterizing pathological tremors.

Main Methods:

• A running second order moment function (SOMF) was developed, calculating moments within a fixed time window.
• EMG data from Parkinson's disease patients and healthy subjects were analyzed.
• Burst detection was achieved by identifying minimum SOMF values below a set threshold.
• Auto- and cross-IBIHs were generated from detected burst intervals.

Main Results:

• The SOMF technique demonstrated high sensitivity and selectivity for detecting EMG bursts, even with significant background noise.
• Tremor frequency and phase relationships derived from IBIHs correlated well with spectral analysis findings.
• The SOMF method proved effective in identifying individual EMG bursts of pathological tremors.

Conclusions:

• The running second order moment function (SOMF) is a suitable method for detecting individual EMG bursts in pathological tremors.
• This non-linear technique is sensitive to non-stationary tremor changes irrespective of amplitude.
• Interburst interval histograms (IBIHs) effectively measure tremor frequency and phase differences.