|
DARPA ITO Sponsored Research
1997 Project Summary | |
| Project Website: | http://www.hgc.cornell.edu/neurons.htm -- Additional project information provided by the performing organization |
| Objective: | The objective of this research is to design, fabricate and exploit arrays of neural cells interfaced to solid state electronics. There is a strong motivation to consider the application of biological neuronal systems interfaced to solid-state silicon-based electronics as specialized computational elements, as sensor elements and as models for distributed signal processing systems. Significant technological and knowledge barriers remain to be addressed to make this a reality. This effort will address the technological issues involved in coupling living neuronal systems and conventional electronics. |
| Approach: | High spatial resolution patterning of surfaces with defined chemistries and topographies will be used to guide the growth and specialization of elements on each neuron. Using new techniques, well-defined biological networks will be directly interfaced with electrode sites at key locations for recording and stimulation. The research effort will carefully study individual neurons and ultimately attempt to develop and study circuits of multiple guided neurons. This will allow unprecedented possibilities in exploring and exploiting the combined computational abilities of organic neuronal systems and silicon based electronics.
Neural networks in the brain are not built in a random way. While it was once popular to view the brain as a random net, research over the last few decades has shown that like a computer chip, brain circuits are assembled according to very accurate blueprints. Mis-wired connections are as deleterious to the function of biological neural networks as they are to electronic circuits. Moreover, unlike the conventional computer chip which is constructed of a relatively small number of different types of components, the brain's circuits are composed of hundreds, or thousands, of different types of neurons, each with distinct physiological properties. This additional complexity is thought to be one of the key factors allowing neural computation in some ways to far surpass that of present conventional computers. This project will employ a strategy that uses the knowledge of the biological controls that underlie the specificity of neuron growth in vivo to produce prescribed networks of neurons. It will control the initial development of neuronal polarityP the orientation of each neuron1s axon and dendrites. It will combine the use of bio-instructive substrates, in this case axon-inducing stimuli, and guidance pathways to direct the axon of one neuron to synapse on another. Moreover, because neurons make only one axon, this approach also controls the orientation of the dendritic tree. Bio-instructive substrates will be fabricated using chemical and topographical modification of silicon substrates. |
| Recent FY-97 Accomplishments: | New Start |
| FY-98 Plans: | The initial phase of the project will be the development of advanced selective surfaces. These experiments will be designed to identify biochemical and topographical surface modifications that will permit guided attachment, growth and orientation of different types of nervous system cells. The response of neurons and astrocytes on coated silicon substrates will be studied. We will consider a range of proteins and other chemicals for controlling the growth and orientation of the neurons and ways of patterning the surface chemistries and topographies. An important aspect will be to determine the best method to orient and control the growth of the axons and dendrites in desired patterns and to develop techniques for multiple layer chemical patterning of surfaces. |
| Technology Transition: | New Start |
| Principal Investigator: |
Harold Craighead School of Applied and Engineering Physics Clark Hall Ithaca, NY 14853 607-255-8707 607-255-7658 fax hgc1@cornell.edu
Michael Isaacson
Daniel Whitaker |