High-Resolution Three-Dimensional Hybrid MRI + Low Dose CT Vocal Tract Modeling: A Cadaveric Pilot Study

Area of Science:

• Medical Imaging
• Anatomy
• Voice Science

Background:

• Magnetic Resonance Imaging (MRI) vocal tract models lack detail in bony structures and have low spatial resolution.
• Computed Tomography (CT) offers better bony detail but involves ionizing radiation risks.
• Current MRI techniques achieve 2 mm³ resolution within 20 seconds, limiting anatomical accuracy.

Purpose of the Study:

• To develop a high-resolution hybrid CT-MRI vocal tract model.
• To combine low-dose CT for bone imaging with accelerated high-resolution MRI for soft tissues.
• To improve the accuracy and detail of vocal tract models for voice research and education.

Main Methods:

• Determined minimum CT dosages and evaluated a custom 16-channel airway receiver coil for accelerated MRI.
• Applied a rigid body landmark-based partial volume registration scheme to blend CT and MRI data.
• Segmented the hybrid CT-MRI model using Slicer software.

Main Results:

• Ultra-low dose CT (0.06 mSv) provided sufficient bone visualization, comparable to transatlantic flight radiation exposure.
• Accelerated MRI achieved 1 mm³ resolution with acceptable image quality from undersampled data.
• Hybrid CT-MRI models demonstrated more detailed 3D vocal tract representations than MRI alone.

Conclusions:

• High-resolution CT and MR image sets can be combined for accurate vocal tract reconstructions, including teeth and bone.
• This hybrid approach enhances anatomical detail for voice production research and education.
• The method advances understanding of vocal tract anatomy and its relation to voice production.