On Self-Organizing Map Based Classification of Insect Neurons

Abstract

In this paper, a systematic method based on selforganizing maps is presented to classify interneurons of silkworm moths. Denseness of branching structures and existence of thick main dendrites are quantified by six fractal dimension values and three values calculated from images to which fundamental processing techniques are applied, respectively. Such values are employed as nine elements in training data for a map. The classification result is obtained as clusters with units in the trained map. Experimental results establish that the classification executed by the proposed method is comparable in accuracy to the manually executed classification.

Reconstruction of virtual neural circuits in an insect brain

“…For example, 153 antennal lobe local interneurons have been registered in the database and classified into five types (Seki and Kanzaki, 2008 ). This classification did correspond to the computational classification by fractal dimension calculated using the box counting method (Urata et al, 2006 ). Anatomical identification and classification have also been done in neuronal populations in the mushroom body, the center of learning and memory, and the lateral accessory lobe, the premotor network in the insect brain, based on 109 and 120 registered neurons, respectively (Fukushima and Kanzaki, 2009 ).…”

Understanding and Reconstruction of the Mobiligence of Insects Employing Multiscale Biological Approaches and Robotics

“…(i) An analysis of 153 local interneurons in the AL that were registered in the BoND revealed that the AL consists of five types of local interneurons [7]. Similar results were obtained by fractal analysis using the box-counting method with a self-organization map [15]. (ii) Principle component analysis of 156 projection neurons (PNs) in the AL revealed that olfactory responses are translated into spatial and temporal patterns in the glomeruli of the AL [12].…”