MIT CogNet, The Brain Sciences ConnectionFrom the MIT Press, Link to Online Catalog
SPARC Communities
Subscriber : Stanford University Libraries » LOG IN

space

Powered By Google 
Advanced Search

 

Modelling the Internal Tongue Using Principal Strains

 Maureen Stone, Danielle Dick, Andrew S. Douglas, Edward P. Davis and Cengizhan Ozturk
  
 

Abstract:

This paper used a newly developed technique, tagged Cine-Magnetic Resonance Imaging ( tMRI), to examine the internal deformation of the tongue during speech. Two studies, each on a single subject, examined local internal tongue motion in 2D and 3D. The first study examined the motion of approximately 40 points within the tongue, in each of three sagittal slices (L, M, R), tracked during the movement from /k/ to /a/. Measurements of tag positions were made at 7 consecutive time-phases in the syllable. A mechanical model of tongue deformation, similar to a finite element analysis, calculated local homogeneous stretch based on the deformations of the tag points between the consonant and vowel. 110 internal tongue locations were probed and principal strains were calculated for them using the model. The principal strains revealed local compression and extension patterns from which inferences could be drawn about the activities of Verticalis, Hyoglossus and Superior Longitudinal muscles. The second study tracked 3D motion and calculated 3D strains from multiplanar tMRI images, for the syllable "sha". Three orthogonal tag planes (x, y, z) were collected for 24 consecutive time-phases in 10 axial and 5 sagittal slices. The sagittal slices were recorded twice, once each with horizontal and vertical tag planes. The axial slices were recorded once with lengthwise (anterior-to-posterior) tag planes. These planes reflect deformations in y, z and x respectively. A B- spline model tracked tag deformation in each plane and reconstructed 3D deformation. This model is being adapted from cardiac tagging methods and preliminary data are presented.

 
 


© 2010 The MIT Press
MIT Logo