| The goal was to develop a detection algorithm for a microbial detection instrument that would, among other things, successfully distinguish between live and dead bacterial cells, bacterial spores, bacterial media, and minimal media/water, with high detection rates and low false alarm rates. It was desired that this occur in real time without any sample contact.; This thesis describes the means by which a detection algorithm was developed that quickly detects bacteria in aqueous solutions with detection limits near 101 cells/cm3 by using the intrinsic fluorescent properties of bacteria. It also describes the data transmission protocol used to transmit information between the PC and the instrument over a potentially lossy channel and describes the software development of the graphical user interface.; The results of the detection algorithm show that it can indeed distinguish between the desired sample groups and lists ideas for further research and development. |
