View Full Version : EMD - electromechanical delay

unknown user
06-18-2005, 07:22 PM

In short answer to Chris' questions:
1. Yes, in fact, there is an electromechanical delay (EMD) at the beginning
(EMAD) and the end of muscle activation (EMRD).
2. The time between the first (EMAD) resp. last (EMRD) appearance of EMG and
the concerning force of a muscle.
3. The value is of about 20 ms for leg and arm-muscles for electromechanical
activation delay (EMAD) and of about 60-100 ms for electromechanical
relaxation delay (EMRD). Additional because of some reasons EMRD is very
difficult to be determined, but is essential for movement control and muscle
coordination resp. inter-muscular-coordination.
I spent a very long time in analysing EMD at the beginning and the end of
muscle activation to get these results.

Please let me give you one important statement concerning the smoothing of
data for your further discussion.
Using absolutely the same input data of several muscles (EMG and force of
MVCs) I compared all methods reported in literature and a lot of threshold
values to determine EMD. The result is a rather distinct dependence of onset
times of both EMG and force on the applied method and within on the
threshold values. But there is no automatic method found providing reliable
and comparable results.
In fact this means there was absolutly no correlation between the results of
two different method using the same input data. In this context a filter
with 2 or 3 Hz distorts the content of EMG and force concerning the rising
characteristic especially in case of very time sensitive data like EMG and
is to be rejected as absolutly unrealistic.
In conclusion for determination of time sensitive values as EMD the visual
method (analyzing EMD by hand) with earliest onset times, lowest threshold
levels and a muscle independent course of EMD related to initial muscle
length seems to be the only advisable regarding accuracy and reliability.
For other cases of motion analysis or motor research with possibly
acceptable inaccuracies of about ±20-30ms the preferable method should use a
moderate low-pass filter and low percentage thresholds.
You mave have a lock to the following page to get these results and to find
some small articles in english language about this theme:
Unfortunately the main publication "The electromecanical delay of human
skeletal muscle" is written as a book in german language:

Best wishes
Mit freundlichen Grüßen

Thomas Jöllenbeck

Priv.-Doz. Dr. Thomas Jöllenbeck
Klinik Lindenplatz
- Institut für Biomechanik -
Orthopädische Rehabilitationsklinik
mit sportmedizinischer Abteilung
Chefarzt Dr. Ch. Schönle
Weslarner Straße 29
D 59505 Bad Sassendorf
Tel.: +49 2921 501-3414
Tel.: +49 2921 501-4326 (Sekr.)
Fax: +49 2921 501-4310
Mobil: +49 170 6524292