A comparative analysis of intraglottal geometry and velocity flow fields: Excised human, canine, and synthetic vocal fold models

Synthetic vocal fold models imitate the characteristics of human vocal folds and offer advantages over tissue models, such as accessibility and prolonged lifespan, all while producing comparable vibration frequencies to those in humans. Nonetheless, due to their simplified design, they may not fully capture the intricate behavior observed in tissue models like the canine larynx model. Canine larynges, like human’s, exhibit a vertical mucosal wave, which yields a phase delay between the inferior and superior edges of the vocal folds free margin. Consequently, a divergent glottis forms during closing, producing intraglottal flow separation vortices (FSVs) that increase voice loudness and intelligibility. To bridge the knowledge gap between silicone and tissue vocal fold models, we compared the intraglottal geometry and velocity flow fields during phonation of these three models. At low subglottal pressures, the synthetic model displayed small divergence angles and no flow separation, where both tissues models exhibited FSVs. At higher pressures, FSVs observed in tissue models were correlated with increased glottal flow waveform skewing and higher MFDR values. These findings highlight why the excised canine larynx model should be the preferred choice over synthetic vocal fold models for studying the aerodynamics and fluid-structure interaction during phonation.