Page 384 - Handbook of Biomechatronics
P. 384

378                                                  Lilach Bareket et al.


             Finally, the “C-sight” project group has demonstrated stimulation of the
          ON using a two-channel penetrating electrode array implanted into the ON
          in rabbits and in cats, and a 16-channel array implanted in rabbits (Chai et al.,
          2008; Li et al., 2009; Yan et al., 2016; Sun et al., 2011). They observed dif-
          ferences in cortical recordings in response to modification of the stimulation
          parameters. For example, it was found that monopolar stimulation with
          three penetrating electrodes induced more localized cortical responses than
          the bipolar stimulation (Cao et al., 2015), and that a charge-balanced
          biphasic pulse with an interphase gap of 0.2ms was the most efficacious
          (Sun et al., 2013). In addition, it was demonstrated that a five-electrode array
          with electrodes placed at different depths in the ON elicited visuotopic elec-
          trical stimulation of the visual cortex with a spatial resolution of about 2–3
          degrees (Lu et al., 2013). Current steering was also used to enhance the spa-
          tial resolution of the stimulation (Yan et al., 2016). The investigators
          suggested that these observations imply that the resolution of an ON visual
          prosthesis may not be limited by the number of electrodes only (Lu et al.,
          2013). Recent investigations also demonstrate the potential of targeting
          the LGN to evoke visual response (Panetsos et al., 2011; Pezaris and
                                                               3
          Reid, 2007). The LGN is a small nucleus (about 250mm volume and
          up to 10mm long) situated in the posterior/inferior part of the thalamus.
          Most of the fibers of the optic tract terminate on neurons in the LGN,
          and it has six distinctive layers of neurons. The visual information from
          the retina to passes through the LGN before reaching the primary visual cor-
          tex. Key features of the LGN are physical segregation, in form of different
          cellular layers, for the magnocellular and parvocellular visual pathways that
          are specific to motion/localization and color/detail respectively and projec-
          tion of retinal structure (Mullen et al., 2008; Wiesel and Hubel, 1966). Also,
          it has a geographic representation of the retina (retinotopic projection). The
          retinotopic projection spans a larger area of tissue (compared with the retina)
          and therefore it is potentially possible to achieve higher spatial resolution for
          similar size electrodes. Beyond these unique features, the relatively small
          dimensions and simple structure of the LGN compared with the visual cor-
          tex, as well as the possibility to apply surgical techniques similar deep brain
          stimulation procedures are additional advantages (Pezaris and Eskandar,
          2009). The LGN could potentially be used as a stimulation target in cases
          where the retina and ON have lost function if the visual cortices have been
          injured or lost (cortical blindness). The disadvantages of the LGN are that
          only a hemifield passes through each side so that bilateral placement would
          be required to cover the entire visual field. Similar to cortical and ON
   379   380   381   382   383   384   385   386   387   388   389