Biomedical Engineering Reference
In-Depth Information
neurophilic and neurophobic patterning. The patterning can be obtained by
chemically depositing permissive hydrophilic layers (organosilanes, laminin,
poly- D -lysine, polyethyleneimine) and non-permissive hydrophobic layers
(polyimide, glass, fluorocarbon) into the desired pattern by lithographic,
deposition or microstamping methods. The final goal is to design circuit
architecture of neural networks with real neurons in the nodes to study the way
real neurons in the network respond to input signals.
However, the reality is the connections made by the neurons are di cult to
control and the function of the network is likely to be compromised. In
addition to blocked ionic channels in the membrane (due to the way the cell sits
on the surface or inside the well), distorted signals may also be generated both
intracellular and extracellular. One possibility to overcome this challenge is to
replace the neurotrophic factors in the implanted electrodes with living neuron
cultures, which will establish dendritic connections to the brain tissue after
implantation, fixing the probe in place and improving the interface with the
nearby neurons.
The most interesting application of in vivo neural networks is the study of the
cell metabolism and the reaction of the cells when exposed to different small
molecules (e.g. pharmacological drugs, neurotrophic factors, neurotoxins) and
other biochemical stimulants of relevant interest (GABA, nerve growth factor,
etc.). In this way the spontaneous oscillatory activity of the living network and
entrainment processes can be studied in greater detail to understand how a
culture of neurons outside the brain organizes itself in such a coherent manner
in the absence of a higher order control signal like the ones supposedly
necessary to coordinate the brain activity in totality.
It has been shown that neurons are extremely sensitive to toxins and that the
cellular death can be witnessed in cultures real time (including the degree of
recovery after the toxin is removed from the system). Furthermore, it is hoped
that these studies will one day render experimental testing with live animals
unnecessary, which would be a remarkable accomplishment.
More importantly, the living neural network has been shown to be trainable
and capable of learning in the same way that an artificial neural network can be
trained to perform patterns of recognition. Such exciting applications are
hoped to one day reveal the brain's cellular mechanisms for learning and
cognitive abilities.
d n 4 t 3 n g | 7
n 3 .
3.9 Microfluidics in Neurobiological Research
Microfluidic devices are emerging as powerful tools in neurobiology because of
their ability to precisely control the spatial and temporal growth of neurons and
manipulate the microenvironment of cells in vitro. Since their introduction in
1998, microfluidics device have been used intensively for cell biology studies.
Miniaturization, integration and automation represent important potential
advantages for exploring neuron cell attachments to substrates, neurite growth,
nerve cell interactions, neural stem cell differentiation, neuropharmacology,
neuroelectrophysiology, and biosensing applications. Although we present
 
Search WWH ::




Custom Search