Noise tolerance as a measure of channel discrimination for multi-channel neural interfaces

Research output: Contribution to journalConference articlepeer-review

1 Scopus citations

Abstract

Spatial selectivity can be obtained when recording the activity of a peripheral nerve [1-4] or corticospinal pathways of the spinal cord [5] by circumferential placement of the metal contacts around the axon bundle. Selectivity indices that have been proposed [1-3], however, do not measure how the channel discriminability deteriorates in the presence of noise. The pattern of amplitude distribution across the recording sites during a neural firing (a vector) can be considered as a symbol received at the end of an information channel. Thus, the performance of the neural interface/recording method can be quantified for the noisy case and compared with others using the classic formulaé for the information channels. Monte Carlo simulations in this study show that the decay of information transfer rate with noise can differentiate between neural interfaces that have identical spatial selectivity indices based on the Euclidian distance measure [2]. Noise tolerance can be the method of choice to assess the performance of multi-channel neural interfaces in terms of channel discrimination.

Original languageEnglish (US)
Pages (from-to)704-706
Number of pages3
JournalAnnual Reports of the Research Reactor Institute, Kyoto University
Volume1
StatePublished - Dec 1 2001
Externally publishedYes
Event23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society - Istanbul, Turkey
Duration: Oct 25 2001Oct 28 2001

All Science Journal Classification (ASJC) codes

  • Energy Engineering and Power Technology
  • Mechanical Engineering

Keywords

  • Information rate
  • Neural interface
  • Selective nerve recording
  • Signal-to-noise ratio

Fingerprint Dive into the research topics of 'Noise tolerance as a measure of channel discrimination for multi-channel neural interfaces'. Together they form a unique fingerprint.

Cite this