Summary: | In this work the use of the central nervous system (CNS) of the aquatic snail Lymnaea stagnalis on planar multi-electrode arrays (MEAs) was developed and analysis methods for the data generated were created. A variety of different combinations of configurations of tissue from the Lymnaea CNS were explored to determine the signal characteristics that could be recorded by sixty channel MEAs. In particular, the suitability of the semi-intact system consisting of the lips, oesophagus, CNS, and associated nerve connectives was developed for use on the planar MEA. The recording target area of the dorsal surface of the buccal ganglia was selected as being the most promising for study and recordings of its component cells during fictive feeding behaviour stimulated by sucrose were made. The data produced by this type of experimentation is very high volume and so its analysis required the development of a custom set of software tools. The goal of this tool set is to find the signal from individual neurons in the data streams of the electrodes of a planar MEA, to estimate their position, and then to predict their causal connectivity. To produce such an analysis techniques for noise filtration, neural spike detection, and group detection of bursts of spikes were created to pre-process electrode data streams. The Kohonen self-organising map (SOM) algorithm was adapted for the purpose of separating detected spikes into data streams representing the spike output of individual cells found in the target system. A significant addition to SOM algorithm was developed by the concurrent use of triangulation methods based on current source density analysis to predict the position of individual cells based on their spike output on more than one electrode. The likely functional connectivity of individual neurons identified by the SOM technique were analysed through the use of a statistical causality method known as Granger causality/causal connectivity. This technique was used to produce a map of the likely connectivity between neural sources.
|