|M.Sc Student||Yekutieli Ziv|
|Subject||Integrated Multi-Electrode Array: An Interface to Ex Vivo|
|Department||Department of Electrical Engineering||Supervisor||Professor Ran Ginosar|
|Full Thesis text|
For more than fifty years, the electro-chemical properties of neurons have been closely scrutinized. The most fundamental procedure in neurophysiology research is probing the membrane of the neuron thus measuring its electrical activity. Having characterized the single neuron very accurately, scientists have shifted their attention to observe the properties and behavior of an assembly of neurons: the neural network (NN). Each and every muscle in our body, our senses, emotions and, basically, our very essence, are governed by NN's in our brain so understanding the NN properties, as well as we understand the single neuron, is a vital step in learning about our brain, treating neurological diseases and curing paralysis.
At the early stages of NN research, the same methods that were used for a single neuron were implemented for the NN simply by adding more and more electrodes to sense more neurons. However, soon those methods were exhausted due to two main reasons: First, there is a physical limitation to the number of electrodes one can place by hand in the very dense area where thousands of neurons are located. Second, when dealing with many neurons rather than just one, new difficulties arise such as how to distinguish between the activities of two neighboring neurons, how to transmit and compute the increasing amount of data and so on.
In order to deal with the first problem, the requirement for increasing the number and density of the electrodes, several methods are developed in order to create a Multi-Electrode Array (MEA) where an ever increasing number of electrodes are placed one next to the other. This work is meant to propose a solution for the second main problem: Given an access to a large number of neurons by means of the MEA, we would like to sample the electrical activity of so many neurons, to distinguish between neurons and to process their activity. In this work we introduce Integrated MEA (iMEA) which is a platform connected to an MEA, that is aimed to provide the requirements listed above with addition of some new tools for making the NN research in the future as accessible as the single neuron research is today.