search

eaertbel / IPSA

Integrated Platform for Spasticity Assessment
2 stars 0 forks source link

Indicating onset period in rmsEMG-position graph #33

Open LynnBaron opened 12 years ago

LynnBaron commented 12 years ago

Be able to manually indicate on-period for current EMG on the rmsEMG position graph.

eaertbel commented 12 years ago

It is already possible to indicate the on-period on the raw EMG graph. Why is it necessary to do this on the rmsEMG graph ? In that case, you are mixing the automatic method (rms + thressholding) with the manual method (starting from raw EMG).

Can you confirm that this is a really necessary feature, or a symptom of featuritis....

LynnBaron commented 12 years ago

featuritis :)

Most of the time I discard noisy raw EMG, but sometimes, I feel that it improved in the rms and decide to keep it. However, it's very difficult to indicate the exact onset on the raw signal of noisy data so I do it on the rms. Because rms is not trustworthy (filters, smoothing...) I want to confirm my onset placement by looking at where the onset occured in relation to the ROM. Artifacts or active participation of the patients will show up at the beginning of the movement. I also use the new graph that you created where I can see onset in relation to position, velocity and torque. but I can only see this graph after I've already indicated the onset. In any case, this is really the least urgent of the issues :)

Also something that is still worrying me a lot: very often EMG onset is still occuring before onset of movement. it's almost impossible that so many children are not relaxed at movement initiation. Is it possible that we have some delay? in the motion graph for instance? I can show you some examples next time we meet.