Abstract

Several functions of daily life depend on us being able to use our voice, whether it is introducing oneself to someone or singing a lullaby. Hence when voice health is compromised it strongly affects our quality of life. Conventional clinical approach to restoring voice health relies heavily on the clinician’s ability to: a) identify aspects of the subject’s behaviour and hygiene e.g. posture, hydration, breath control and b) assess the deviation in each aspect from a poorly characterized reference. Therefore it is critical to develop an objective, mechanistic description of how the mechanics of voice production affects voice health.

Voice is produced when air pressure difference along the length of the airway causes the pair of vocal folds to vibrate and collide with each other at high frequencies (~ 200 Hz). Past computational models failed to bring together features of this process that are important for the determination of mechanical stresses within the vocal folds. These include three-dimensionality, fluid–structure interaction, self-sustained oscillations, realistic vocal fold constitutive properties, contact interaction and surface adhesion. In this talk, I will present a validated computational model that for the first time overcomes each of these challenges in the same model. Model predictions led to the discovery that vibration and collision induced stresses give rise to non-negligible rates of dehydration in critical regions of vocal fold tissue. Vocal fold surface adhesion characteristics were found to selectively control post-collision separation characteristics as observed in high-speed imaging experiments (e.g. formation of fluid bridges). Furthermore, these results were analysed within a patient-specific framework to aid clinical decision-making. We were thereby able to establish an objective mechanistic basis for the influence of voice production on voice health and to advance the state-of-the-art in in silico management of voice disorder.