Reduced order modeling of passive and quasi-active dendrites for nervous system simulation Academic Article uri icon

Overview

MeSH Major

  • Action Potentials
  • Dendrites
  • Models, Neurological
  • Nerve Net
  • Neural Networks (Computer)

abstract

  • Accurate neuron models at the level of the single cell are composed of dendrites described by a large number of compartments. The network-level simulation of complex nervous systems requires highly compact yet accurate single neuron models. We present a systematic, numerically efficient and stable model order reduction approach to reduce the complexity of large dendrites by orders of magnitude. The resulting reduced dendrite models match the impedances of the full model within the frequency range of biological signals and reproduce the original action potential output waveforms.

publication date

  • October 2011

Research

keywords

  • Academic Article

Identity

Language

  • eng

Digital Object Identifier (DOI)

  • 10.1007/s10827-010-0309-5

PubMed ID

  • 21225333

Additional Document Info

start page

  • 247

end page

  • 71

volume

  • 31

number

  • 2