Neural Network Facilitated Analysis of Stress in a Biological Species

E.E. Udoh, D. Erbach, and A. Mustafa (USA)

Keywords

Data mining, neural networks, stress prediction, and biological species

Abstract

Artificial neural networks (ANN) are becoming a common tool for modeling complex input-output dependencies. This paper describes the mining of biologically generated data using a neural network application, BrainMaker. In a period of ten weeks, the fish species (steelhead trout) was monitored under normal (50 55 Fahrenheit) and abnormal (60-70 Fahrenheit) temperature conditions. Biological parameters such as weight, length, glucose, packed cell volume (PCV), phagocytosis and cortisol were measured and stored in a database. The objective of this work is to predict the level of cortisol, a chemical in the fish, that could be correlated with the stress level in fish. The work shows that the future levels of cortisol or the stress placed on the fish with certain traits can be predicted using ANN.

Important Links:



Go Back


IASTED
Rotating Call For Paper Image