Semi-automatic segmentation of speech articulators using MRI images

Yuanming Suo, Emi Z. Murano, Maureen Stone, Paul M. Evitts, Jerry L. Prince

Research output: Contribution to conferencePaper

Abstract

This paper used the random walker, a multiple structure semiautomatic segmentation approach, for rapid delineation of speech articulators in magnetic resonance imaging (MRI) images. Results on the delineation of the tongue and surrounding structures in both 2D and 3D are shown. Efficiency and sensitivity of the approach as well as other factors that may affect the results were assessed by analyzing the delineations of naive raters who used the program. The results of both naive raters and experts using the semi-automatic program were compared to expert manual delineations. Our results show that this semi-automatic approach is efficient, accurate, and easy to use.

Original languageEnglish (US)
Pages9-15
Number of pages7
StatePublished - Jan 1 2011
Event2011 9th International Seminar on Speech Production, ISSP 2011 - Montreal, Canada
Duration: Jun 20 2011Jun 23 2011

Other

Other2011 9th International Seminar on Speech Production, ISSP 2011
CountryCanada
CityMontreal
Period6/20/116/23/11

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language
  • Computer Science Applications

Fingerprint Dive into the research topics of 'Semi-automatic segmentation of speech articulators using MRI images'. Together they form a unique fingerprint.

  • Cite this

    Suo, Y., Murano, E. Z., Stone, M., Evitts, P. M., & Prince, J. L. (2011). Semi-automatic segmentation of speech articulators using MRI images. 9-15. Paper presented at 2011 9th International Seminar on Speech Production, ISSP 2011, Montreal, Canada.