Abstract
Introduction: It has previously been reported(1,2,3) that EMG signals from the lumbar spine are highly prone to contamination by ECG artefacts. As the ECG spectrum overlaps an area of interest in the EMG spectrum this has obvious implications for the accurate analysis of EMG data.
Methods: EMG data was recorded from 192 subjects across two years (initial contact, 12 months and 24 months). When a moving average filter was applied to this raw data an obvious ECG trace could be observed in the case of a large proportion of the tests. The application of a Fast Fourier Transform on this raw data demonstrated a large low frequency spike, with little known correlation to lumbar muscle spectral characteristics, but highly indicative of an ECG signal.
As multiple source signals were recorded per test, the Independent Component Analysis technique was able to be used to split the EMG raw signal into statistically independent components. This technique is designed to take the multiple signal inputs, and convert them into multiple outputs, where the inputs are distinguishable by electrode location; the outputs are distinguishable by signal biological origin.
Results: Upon extraction, one of the signal traces showed a clear ECG trace. The Fourier Transform of this trace showed the low frequency spike, with no other signal components present. The Fourier Transform of the EMG trace showed the original EMG graph, with no low frequency peak. Specific spatial information has been exchanged for a much cleaner signal.
Conflicts of Interest: None
Source of Funding: None
Correspondence should be addressed to: SBPR at the Royal College of Surgeons, 35–43 Lincoln’s Inn Fields, London WC2A 3PE, England.
References
1 Allison GT. Trunk muscle onset detection technique for EMG signals with ECG artefact. J Electromyogr Kinesiol2003;13:209–216. Google Scholar
2 Mathieu PA, Fortin M. EMG and kinematics of normal subjects performing trunk flexion/extensions freely in space. J Electromyogr Kinesiol2000;10:197–209. Google Scholar
3 Panagiotacopulos ND, Lee JS, Pope MH, Friesen K. Evaluation of EMG signals from rehabilitated patients with lower back pain using wavelets. J Electromyogr Kinesiol1998;8;269–278. Google Scholar
