15th International Congress of Phonetic Sciences (ICPhS-15)
Tracking the motion of the tongue, both its surface and its bulk, is important in understanding how the tongue muscles contribute to speech production and control. In order to understand the motion in the interior of the tongue, tagged magnetic resonance imaging (tMRI) has been used. Tags are distinctive features on tMRI images that can be tracked to calculate various indices of motion. The present methods of tracking tags and extracting motion information are manually intensive and time-consuming, however, typically leading only to coarse estimates of the tongues motion. In this work, we have used harmonic phase magnetic resonance imaging (HARP-MRI) for measuring the motion of the tongue from sinusoidally tagged MR images. HARP-MRI uses Fourier filtering and special processing algorithms to calculate several measures of motion like strains, displacements, and motion tracks of tissue points. From these dense measures of tongue motion, inferences can be drawn on the corresponding muscle activity, which is important to understand speech control and related disorders.
Bibliographic reference. Parthasarathy, Vijay / NessAiver, Moriel / Prince, Jerry L. / Stone, Maureen (2003): "Tracking tongue motion from tagged magnetic resonance images using harmonic phase imaging (HARP-MRI)", In ICPhS-15, 2917-2920.