Development and evaluation of synthetic vocal fold phantoms for anisotropic behavior

Abstract

This thesis presents the design, simulation, fabrication, and evaluation of a novel synthetic vocal fold phantom with tunable and anisotropic mechanical behavior, developed to address existing limitations in vocal fold modeling. The motivation for this work stems from the need to create a reusable and physiologically relevant model that can replicate the direction-dependent stiffness of human vocal folds while allowing tunable mechanical responses. Most existing synthetic models are either isotropic, composed of multilayered or fiber-reinforced structures that are complex to fabricate, or are made from hydrogel-based materials that are prone to dehydration and degradation. In contrast, this study introduces an accessible, durable, and experimentally flexible solution using AKSIL®️ silicone phantoms with engineered geometries and internal configurations. In this study, two vocal fold phantoms were developed. The first phantom featured a soft cubic platinum-cure RTV silicone base with a side of 15 mm, which represents vocal fold dimensions, embedded with a central cylinder made of a stiffer silicone to passively induce anisotropy via material contrast. The second phantom was more advanced, and it incorporated a natural latex rubber tube filled with varying water-to-air ratios. This phantom enabled post-fabrication tunability of stiffness levels in the longitudinal direction. Both phantoms were first modeled using COMSOL Multiphysics with appropriate boundary conditions. The selected physics module was Solid Mechanics for mechanical testing. By conducting physical experiments, including uniaxial compression in lab settings, the models were then validated. In addition to static mechanical testing, a vibratory test was conducted using a one-layer vocal fold structure subjected to pulsatile airflow. The latex-filled phantom was mounted in a custom experimental setup that simulated phonatory conditions, and the vibration was recorded using a high-speed CMOS camera. The frequency response of the phantom was evaluated for an internal fluid content of 60% water. The measured fundamental frequency was 101 Hz, which falls within the typical human phonation range, supporting the phantom's dynamic validity and alignment with expected phonatory behavior. Experimental results demonstrated that the first phantom showed different mechanical behavior in longitudinal and transverse directions because of the stiffer silicone core, with Young's modulus values of 21.77 kPa in the longitudinal direction and 3.94 kPa in the transverse direction. The correlation between the simulation and experimental results demonstrates that the proposed method for introducing anisotropic behavior to a vocal fold model is reliable. However, this phantom lacked tunability and required full re-fabrication for each configuration. The second phantom, by contrast, offered both anisotropy and tunability: the Young's modulus in the longitudinal direction ranged from 12.18 kPa (0% water) to 47.34 kPa (80% water), while the transverse modulus remained constant (~2.7 kPa). The same cube was reused across all tests, and water-to-air ratios were modified incrementally, making this method cost-effective, time-efficient, and structurally consistent. The vibratory test conducted on the second phantom confirmed that the phantom is capable of reproducing a realistic vibratory response within the human phonation range. The novelty of this study is the integration of directional stiffness and post-fabrication tunability within a single, simplified vocal fold phantom using accessible and inexpensive materials. The mechanical behavior exhibited by the model is physiologically relevant, and its reusability and modularity make it an ideal candidate for studying phonation mechanics, vocal pathology, prosthetic development, and pre-clinical testing. This study bridges a significant research gap in vocal fold modeling and also provides a foundation for future exploration of anatomical shaping, real-time stiffness modulation, and integrated sensing technologies.