Towards Dynamic 3D MRI of Speech

Yinghua Zhu¹, Yoon-Chul Kim¹, Michael Proctor¹, Shrikanth Narayanan¹, Krishna Nayak¹

MRI has a long history of informing basic research into speech production, and has included the use of static and dynamic 2D imaging and static 3D imaging. In this work, we report on our initial efforts towards real-time 3D imaging of speech using 3D stack-of-spirals gradient echo sequences at 1.5T. We characterize the trade-off between spatial and temporal resolution, and assess the image quality in the articulators (i.e. air-tissue boundaries in the tongue, lips, and velum) using a variety of the spiral readout durations and number of interleaves.