Estimating nonlinear spatiotemporal membrane dynamics in active dendrites

Toshiaki Omori

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)

Abstract

Recent advances in measurement technology enables us to obtain spatotemporal data from neural systems as imaging data. In this study, we propose a statistical method to estimate nonlinear spatiotemporal membrane dynamics of active dendrites. We formulate generalized state space model of active dendrite, based on multi-compartment model. Membrane dynamics and its underlying electrical properties are simultaneously estimated by using sequential Monte-Carlo method and EM algorithm. Using the proposed method, we show that nonlinear spatiotemporal dynamics in active dendritic can be extracted from partially observable data.

Original languageEnglish
Pages (from-to)27-34
Number of pages8
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8834
Publication statusPublished - 2014 Jan 1

Keywords

  • Dendrite
  • Multi-compartment model
  • Probabilistic time-series analysis
  • Spatiotemporal dynamics

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Estimating nonlinear spatiotemporal membrane dynamics in active dendrites'. Together they form a unique fingerprint.

Cite this