Subject-Specific Computational Modeling of Evoked Rabbit Phonation

Area of Science:

• Biomechanics
• Computational Biology
• Acoustic Science

Background:

• Developing accurate computational models for individual voice production is challenging due to unknown vocal fold parameters and validation difficulties.
• Subject-specific modeling requires precise characterization of vocal fold tissue properties.

Purpose of the Study:

• To explore methods for determining individual-specific vocal fold properties using a hybrid computational model.
• To validate a 3D/1D computational model against experimental data for rabbit phonation.

Main Methods:

• Coupling a 3D vocal fold model (based on MRI scans) with a 1D glottal airflow model for efficient simulations.
• Utilizing evoked rabbit phonation and individual subject measurements to determine vocal fold mechanical properties.
• Comparing simulation results (vibration frequency, deformation amplitude) with experimental data.

Main Results:

• The hybrid 3D/1D model successfully simulated rabbit vocal fold dynamics.
• Model parameters were determined to match individual subject's vibration frequency and deformation amplitude.
• Reasonable agreement was achieved between model predictions and experimental measurements.

Conclusions:

• The developed hybrid modeling and validation approach is effective for determining subject-specific vocal fold properties.
• This methodology can advance the creation of personalized computational models for voice production.
• The study provides a framework for improving high-fidelity voice production simulations.