Dear All,

I think this discussion of the technical aspects of linear envelope (LE)
processing of EMG is very interesting. However, it is important to not lose
sight of the purpose of calculating the LE. It is designed to emulate the
force development over time in the muscle. Thus, the low pass filter
cut-off should be such to match the twitch characteristics of the muscle
being studied. I believe that Winter has some data in one of his textbooks
(sorry, can't recall which one at this point in time) that gives examples of
appropriate cut-offs for slow twitch, fast twitch and mixed muscle. The
range was around 1 to 2.5 Hz, I think. This would seem to suggest that 10
Hz in the original email would be far too high. Also, if one is using, for
instance, a Butterworth-type filter, it should be a second-order, critically
damped one. In the end, with the appropriate choice of cut-off and damping,
you should be able to have a reasonable estimate of the shape of the muscle
force-time curve and it will account for the electro-mechanical delay as


Drew Smith, PhD

Assistant Professor
Department of Rehabilitation Sciences
The Hong Kong Polytechnic University
Hung Hom, Kowloon
Hong Kong SAR

Voice: +852 2766 7094
FAX : +852 2764 1435

To unsubscribe send SIGNOFF BIOMCH-L to
For information and archives: