Pattern Recognition System Based on Data Mining for Analysis of Chemical Substances in Brain

Abstract

This paper presents a data mining system for analyzing biochemical changes in the brain of rodents. Manual analysis of such experiments is impractical due to a huge volume of generated data and tedious analytical procedures; as a result, important information is lost. Addressing this issue, our paper proposes a data mining system consisting of several steps (pre-processing, data classification, etc.). In some of the steps we apply the artificial neural network based on the adaptive resonance theory. This paper describes the proposed system and experiments performed to validate it. In the experiments, glutamate and aspartate neurotransmitters in samples extracted from rodent brains were analyzed.