Recording from dorsal roots - problems...

Partap S Khalsa pkhalsa at wpi.WPI.EDU
Thu Nov 4 10:29:00 EST 1993


In article <1993Nov3.224158.26720 at ee.surrey.ac.uk> D.Banks at ee.surrey.ac.uk (Danny Banks) writes:
>
>I am trying to make a good case for using an array of highly selective 
>electrodes for recording sensory signals from the spinal nerves; primarily
>the dorsal roots. This being preferable to using less selective electrodes
>implanted much more distally to the spinal cord to record from nerve
>branches innervating specific areas of the body. The eventual application
>would be in sensory feedback for FES in cases of spinal cord injury.
> . . .

Danny,
  I have no experience with implantable electrodes, but I have done a modest
amount of recording from mechanoreceptors in dorsal roots (cats). To obtain
a clean signal usually requires physical isolation of the specific 'rootlet'
from which you wish to record. We do this in our lab by physically 
teasing apart the dorsal root into very small rootlets (this procedure is
done under an operating microscope). Even with these very fine rootlets, if
we have not previously denervated the surrounding regions (i.e. where we
don't wish to record from), our signal is uninterpratable. With denervation
(transection of other peripheral sensory nerves), if we are lucky, we can
get a clean signal by using a hardware window discriminator. Alternatively,
a software cluster analysis could also be employed.
  As such, I'm sceptical of using an implantable electrode to obtain clean
recordings.

Partap S. Khalsa, MS, DC, FACO
Post-Doc. Research Fellow



More information about the Neur-sci mailing list