Automatic Identification of Phasic Dopamine Release

The study and analysis of dopamine (DA) release in the organism are of great importance due to the fact that this neurotransmitter directly influences processes such as cognition and drug abuse. The fast scan cyclic voltammetry technique allows efficient recording of the phasic release of DA. However, due to the high temporal resolution of the technique, the experiments generate large amounts of data, resulting in a slow and repetitive manual analysis. This paper aims to present and evaluate the performance of a baseline automatic identification system starting from phasic dopamine release images, using different texture descriptors, and combining them searching for complementarity to improve the system overall performance. The best obtained accuracy using automatically extracted patches was 89.18%, and the best f-measure was 77.23%. Also, the DA release dataset used in this work is publicly available at https://web.inf.ufpr.br/vri/databases/phasic-dopamine-release/.