Objective:
MRI has emerged as a powerful modality for vocal tract imaging due to its non-invasiveness and excellent soft tissue contrast. MRI based vocal tract models is used in studying the vocal tract area functions and voice quality1-5 and physiological changes during vowel production6-8.Bony structures relevant in speech production exhibit short spin-spin relaxation (T2) time constants which causes the signal from these structures to rapidly decay after excitation. MRI sequences with echo-times greater than even 1ms are insensitive to image signal from bony structures. With zero echo time (ZTE) MRI techniques the gradient pulses are applied before the excitation pulses enabling imaging with “zero-echo-time" and have demonstrated utility in imaging bony structures. The objective of this work is to evaluate the utility of ZTE in imaging bony structures relevant in voice-production and validate it against ultra-low-dose Computed Tomography.

Methods:
A total of 15 subjects were recruited and scanned with upper-airway ultra-low dose CT and vocal tract MR imaging protocols in accordance with the University of Iowa’s IRB protocol (IRB # 202202556). Ultra-low dose CT was performed with the following scan settings: scan rotation time of 0.25 second and kVp of 120; radiation dose of 0.06 msV. ZTE scans were performed on a GE 3 Tesla Premier scanner with a 21-channel head-neck coil with the following specifications: field of view: 24 cm x 24 cm x 12 cm; resolution: 0.8_0.8_0.9mm; flip angle = 1 degree, receiver bandwidth = 62.50 Hz/pixel; scan time = 11.5 mins; Number of averages = 4. ZTE scans were bias corrected9 and registered with CT counterparts. A 2D Histogram distribution of the registered ZTE and CT pixel intensities in the teeth and jaw region (combined) and the hyoid bone were plotted. Linear regression analysis was used to correlate relationship between the pixel intensities.

Results
We observed strong correlations between ZTE derived pixel intensities and CT based Hounsfield units, suggesting ZTE to be a promising approach to safely image bones without ionization.

Conclusions
Future work will include applying non-rigid registration methods to merge bony structures from ZTE into dynamic speech scans to improve modeling the dynamics of upper-airway.