[22a-A106-1~8] 12.6 Nanobiotechnology

△ [22a-A106-7] Analysis of Connection Strength between Neurons Using Joint Entropy

〇Wataru Minoshima^1,2, Yumi Segawa¹, Chie Hosokawa¹ (1.Osaka Metropolitan Univ., 2.NICT)

Keywords:Cultured Neuronal Network, Multisite Extracellular Potential Measurement, Information Theory

Information processing in the brain is based on the spatiotemporal dynamics of neuronal activity based on the synaptic connections. In measuring extracellular potentials using multi-electrode arrays (MEAs) and spike-train analysis, spatial transmission properties through synapses have been insufficient. We applied transfer entropy in the spike train of neuronal activity measured by multi-electrode arrays to evaluate the connection strength between neurons. However, transfer entropy cannot be applied in real-time evaluation of neuronal connection strength due to its computational load. Here, we analyzed the connection strength of spikes using joint entropy. As a result, connection strength changes can be evaluated by joint entropy in development of neuronal networks. These results suggested that the joint entropy can be used in the real-time analysis of neuronal connection strength.