Incorporating Dental Anatomy Into MRI-Based Vocal Tract Models Using Zero Echo Time Imaging

Area of Science:

• Medical Imaging
• Speech Science
• Biomedical Engineering

Background:

• Accurate vocal tract modeling requires detailed representation of soft tissues, teeth, and mandible.
• Conventional MRI struggles to image teeth (short-T2 tissues), while CT involves ionizing radiation.
• Existing MRI-based vocal tract models often exclude dental structures, limiting anatomical completeness.

Purpose of the Study:

• Evaluate zero echo time MRI (ZTE-MRI) for radiation-free imaging of dental and mandibular structures.
• Assess the impact of incorporating ZTE-derived teeth into MRI-based vocal tract models.
• Compare geometric and acoustic outcomes of hybrid models with conventional approaches.

Main Methods:

• Twenty-five participants underwent ultra-low-dose CT and MRI, including ZTE-MRI and conventional GRE sequences.
• Dental structures were segmented from ZTE-MRI and CT, then fused with GRE-derived airways.
• Hybrid ZTE+GRE-MRI and CT+GRE-MRI models were constructed; acoustic simulations were performed for three voice users.

Main Results:

• ZTE-MRI clearly visualized teeth and mandible with pseudo-CT contrast, handling dental hardware better than CT.
• Hybrid ZTE+GRE models closely matched CT-based models (1.7-2.3% area difference).
• GRE-only models overestimated oral airspace (up to 16-17%), causing measurable acoustic shifts, especially in formant frequencies.

Conclusions:

• ZTE-MRI enables anatomically complete, radiation-free vocal tract modeling including dental structures.
• ZTE-derived teeth yield geometric and acoustic effects comparable to CT, addressing a key MRI limitation.
• This offers a practical, radiation-free alternative for speech and voice research.