AnyBody modeling result vs. EMG study of neck muscles

AnyBody support,

My research team used EMG to evaluate the loading of the cervical spine with overhead pushing and pulling tasks before.

30 healthy subjects performed isometric overhead pushing and pulling tasks with their neck in neutral postures, exerting 25%, 50%, and 75% of their maximum isometric strengths. The activities of the major neck muscles, the sternocleidomastoid and the upper trapezius, in the cervical region (C4-C5 level) were evaluated bilaterally using electromyography (EMG) and expressed in terms of a normalized mean absolute values (N-MAV).

During overhead pushing, the N-MAV of the sternocleidomastoid increased from 5.96 (±4.78) to 10.3 (±7.97) to 16.6 (±11.5) with an increase in the exertion level from 25% to 50% to 75%. The corresponding values during pulling increased from 5.15 (±4.66) to 5.32 (±4.35) to 6.93 (±6.20).

But when we tried to use AnyBody to simulate the overhead pushing and pulling tasks, we didn’t find any magnitude of the activity and Fm for the muscles on the neck, especially for the sternocleidomastoid muscle, for both overhead pushing and pulling tasks.

Will you please offer us help or suggestion for solving this problem?


Hi Cathy,

We actully found a bug in the muscle definition thanks to your comments. It appears that the insertion point of the sternocleidomastoid that should be on the clavicle is on the thorax instead. That definitly has an influence on his activation.
We will correct this in the next repository release, meanwhile i invite you to move the insertion point (ThoraxSeg.SternocleidomastoidClavicleC0NodeR) to the Clavicula segment. That should improve your results.

Best regards, Sylvain.

Hi, Sylvain,

Thanks for your input. It is greatly appreciated!!

After I moved the insertion point to the clavicula segment, the activity of Sternocleidomastoid was indeed better.

I found a problem with the model “overloaded muscle configuration”, even I changed the tEnd from 1 to 4 seconds, and changed the shoulder movements to be constant values in the external text file. You know, I only exerted 75% loads of the maximal strength of the subjects on the model, I also scaled the body weight and height, using ‘scalinguniform’ and ‘anymanuniform’.

When I increased SpecificMuscleTensionShoulderArm= 90 to 120, there are still overloaded muscle configuration. when I tried with 150 value, there was no overloading, but the activity of Sternocleidomastoid went back to almost 0.

Could you please help me with this muscle overloading problem?
Thanks a lot!


Hi Cathy,

The first thing to do would be to check the boundary conditions of the real experiment. For example if the subject was grabing a bar or some kind of handles while pushing on the mesuring device, that provides reactions (moments) that the body can use. And this should lower the muscle activity.
At the moment in your model the contact force is realy modeled as a single point contact with no moments. So make sure you are not missing any component of the contact forces.

Another point is that we also noticed that for large glenohumeral flexion or abduction movement the model has a tendency to be a little weaker than expected, so in this case it is ok to increase sligthly the SpecificMuscleTensionShoulderArm.

Best regards, Sylvain.