The vocal folds are essential for speech production, and a better understanding of vocal fold vibration characteristics may help improve treatments of voice disorders. However, studying real vocal folds presents significant challenges. In-vivo studies are limited by access and safety issues. Excised larynges have a short useable lifetime (on the order of minutes) and are difficult to parameterize. In contrast, synthetic vocal fold models have long useable lifetimes and can be easily parameterized. In this thesis, a series of tests performed on recently developed synthetic, self-oscillating models of the human vocal folds are discussed. These tests include measurements of vibration frequency, sub-glottal pressure, and time-averaged flow rate. The differences between one-layer and two-layer synthetic models are evaluated. Comparisons are made between synthetic model and real vocal fold behavior. The synthetic model is shown to have vibrated at frequencies, pressures, and flow rates consistent with human phonation. The influence of sub-glottal tube length on model vibration frequency is examined. Motion is observed using high-speed imaging. Velocity measurements of the glottal jet using particle image velocitmetry (PIV) were performed with and without an idealized vocal tract, including the effects of the false folds, for various cases of vocal tract asymmetry. Glottal jet velocities measured using PIV were consistent with velocities measured using excised larynges. A starting vortex was observed in all test cases. The presence of the false folds acted to restrain the sides of the starting vortex, and in some cases created new vortical structures shed from the false folds. An algorithm was created to calculate and visualize the jet core centerline. In the vocal tract cases, the glottal jet tended to skew toward the nearest wall; in the false fold cases, the opposite trend was observed as the jet skewed away from the nearest wall (towards the midplane). Plots of RMS velocity showed distinct regions of shear layer and jet core. Vocal tract cases at pressures much greater than phonation onset pressure showed significant increases in RMS velocities compared to open jet and false fold cases.

This study examined the differences between normal vocal fold vibration and the movement patterns of vocal folds with mass lesions by means of a synthetic model. The experimenter molded and cast three sets of vocal folds, representing normal structure, small nodules, and larger nodules. Acoustic, aerodynamic, and digital video signals were recorded and analyzed in order to quantify air flow and pressure, measure vibratory stability, and visually assess closure patterns across the three structural conditions. Statistical analysis revealed that the presence of vocal nodules resulted in a significantly higher onset pressure, fundamental frequency, airflow at onset, and offset pressure. However, the results were inconclusive with regard to vocal stability, and it remains unclear whether the current models of nodules are sufficiently similar to the human system to adequately model the type of mass lesions typically seen in a clinical context.

The voice plays a vital role in human communication. The purpose of voice research is to advance the understanding of voice production physics, with the ultimate goal of leading to improved voice care. In this research computational and synthetic vocal fold models were used to explore the role of subglottal geometry in vocal fold vibration. Three specific studies were performed. First, the effect of the inferior vocal fold surface angle on voice production was investigated using a two-dimensional self-oscillating finite element vocal fold model. Varying the inferior angle resulted in significant changes to model vibratory motion, glottal width, flow rate, and energy transfer. The changes were attributed primarily to changes in structural, rather than aerodynamic, factors. Second, subglottic stenosis (SGS) was introduced and parametrically varied in a similar computational model to determine the influence of SGS on vocal fold vibration. High severities of SGS influenced several factors related to vibration, including glottal width, flow rate, flow resistance, and vibration frequency. Subglottal pressure distributions and flow patterns were also affected. Third, the response of a self-oscillating silicone vocal fold model to varying degrees of SGS in an experimental setup was studied. Consistent with the computational SGS study, SGS had an effect on the synthetic model response at high severities. Changes were seen particularly in subglottal pressure and radiated acoustic sound, and consequently glottal efficiency, which may have important implications regarding the effect of SGS on the human voice.

Fluid structure interactions occur to some extent in nearly every type of fluid flow. Understanding how structures interact with fluids and visa-versa is of vital importance in many engineering applications. The purpose of this research is to explore how fluids interact with flexible and rigid structures. A computational model was used to model the fluid structure interactions of vibrating synthetic vocal folds. The model simulated the coupling of the fluid and solid domains using a fluid-structure interface boundary condition. The fluid domain used a slightly compressible flow solver to allow for the possibility of acoustic coupling with the subglottal geometry and vibration of the vocal fold model. As the subglottis lengthened, the frequency of vibration decreased until a new acoustic mode could form in the subglottis. Synthetic aperture particle image velocimetry (SAPIV) is a three-dimensional particle tracking technique. SAPIV was used to image the jet of air that emerges from vibrating human vocal folds (glottal jet) during phonation. The three-dimensional reconstruction of the glottal jet found faint evidence of flow characteristics seen in previous research, such as axis-switching, but did not have sufficient resolution to detect small features. SAPIV was further applied to reconstruct the smaller flow characteristics of the glottal jet of vibrating synthetic vocal folds. Two- and four-layer synthetic vocal fold models were used to determine how the glottal jet from the synthetic models compared to the glottal jet from excised human vocal folds. The two- and four-layer models clearly exhibited axis-switching which has been seen in other 3D analyses of the glottal jet. Cavitation in a quiescent fluid can break a rigid structure such as a glass bottle. A new cavitation number was derived to include acceleration and pressure head at cavitation onset. A cavitation stick was used to validate the cavitation number by filling it with different depths and hitting the stick to cause fluid cavitation. Acceleration was measured using an accelerometer and cavitation bubbles were detected using a high-speed camera. Cavitation in an accelerating fluid occurred at a cavitation number of 1.

Voice is an important tool for communication. Consequently, voice disorders tend to severely diminish quality of life. Voice research seeks to understand the physics that govern voice production to improve treatment of voice disorders. This thesis develops a method for creating complex synthetic laryngeal models and obtaining flow data within these complex models. The method uses Computed Tomography (CT) scan data to create silicone models of the larynx. Index of refraction matching allows flow field data to be collected within a synthetic complex larynx, which had previously been impossible. A short proof-of-concept of the method is set forth. Details on the development of a mechanically-driven synthetic model are presented. Particle image velocimetry was used to collect flow field data in a complex and a simplified supraglottal model to study the effect of complex geometry on the supraglottal jet. Axis switching and starting and closing vortices were observed. The thesis results are anticipated to aid in better understanding flow structures present during voice production.

The influence of vocal fold (VF) vibration on perfused flow through VF vasculature is an area of research that has previously received limited attention. The aim of the research presented in this thesis was to contribute towards an improved understanding of the effects vibration on perfusion through vasculature within the VFs. This was done using a series of computational simulations of geometric changes to, and perfusion through, microchannels embedded in VF models. A computational structural model based on synthetic VF models used in previous experimental studies was first developed. The model and its embedded microchannel were initially studied under static pressure loads applied to the inner surfaces of the channel as well as to the VF inferior and medial surfaces. It was shown that the channel volume decreased linearly and the channel length increased quadratically with increasing pressure on the external VF surfaces. Changes in Poisson's ratio and its influence on the embedded channel's maximum deflection, volume, and length were also studied. Across the range of Poisson's ratios that has been studied for silicone used in synthetic VF models (0.4 to 0.495) there was shown to be limited change in microchannel maximum deflection, channel length, and volume for equivalent pressure loads. The model was then modified to include an external oscillating pressure load on the VF surface that caused the model to vibrate. Two separate studies were conducted to determine how frequency and deflection amplitude affected the predicted perfusion flow rate through the embedded microchannel by accounting for the changes in microchannel geometry during vibration. These studies showed that frequency had little effect on predicted flow rate, while increased deflection amplitude led to greater reductions in predicted flow rate. These reductions in flow rate were attributed to channel lengthening and cross section deformation during vibration, with the latter playing a much larger role. Reductions in flow rate results were found to favorably agree with measured experimental flow rate reductions reported previously. Computational fluid dynamics simulations of water flowing through the inflated embedded microchannel during vibration were also conducted. These simulations were used to explore how changes in vibration length, amplitude, and frequency affected the fluid dynamics in the microchannel whilst minimizing geometric changes to the microchannel. The flow rates from each of the simulation cases were compared to determine which of the vibration parameters contributed the most to flow losses. Vibration length and amplitude were shown to be statistically significant. An investigation was undertaken to further elucidate the mechanisms behind the flow losses induced by vibration. The effects of channel elongation, increased channel curvature, pressure rises, and pressure gradients during vibration were analyzed. Changes in channel elongation and pressure were shown to significantly contribute to flow losses and flow rate reduction. The results from these simulations were compared with the structural simulations which analyzed how changes in microchannel geometry affected flow rate reductions. Changes in the microchannel geometry were shown to contribute much more significantly to reductions in perfusion flow rate compared to changes in vibration parameters (i.e., vibration length, amplitude, and frequency).

No description available.

No contexto da fonação, as pregas vocais são responsáveis pela produção primária do sinal de voz na laringe. Tão importantes quanto complexas, elas exibem uma dinâmica rica e uma série de comportamentos que determinam as características fundamentais do sinal emitido pela boca. Com o intuito de avaliar suas propriedades sob a luz dos métodos da dinâmica não-linear de sistemas, este trabalho tem por objetivo principal desenvolver uma técnica de extração e simulação de sua dinâmica. Para isto são empregados modelos determinísticos das pregas vocais e imagens de videolaringoscopia de alta velocidade. Estes elementos são usados para reproduzir a dinâmica e cinemática das pregas de indivíduos saudáveis através de um procedimento de otimização dos parâmetros mecânicos destes modelos de pregas. Feito isto, os modelos são usados para avaliar o comportamento não-linear das pregas vocais através de uma análise modal não-linear cujo propósito foi descrever como as não-linearidades estruturais das pregas podem influenciar na voz. Os resultados mostram a eficiência de um método capaz de reproduzir a maior parte dos padrões de vibração das pregas e uma relação de dependência entre sua energia mecânica e sua frequência fundamental de vibração, que é fruto das não-linearidades estruturais das pregas vocais. / As far as phonation is concern, the vocal folds produce the primary voice signal within larynx. As important as complex, they exhibit a rich dynamics capable to influence the fundamental voice features. In light of nonlinear dynamical methods, this work has as purpose to develop a method to extract and simulate the vocal folds dynamics. Deterministic vocal folds models and high-speed videolaryngoscopy are used to reproduce the dynamics of healthy subjects and evaluate their nonlinear behavior and some phenomena raised by the nonlinearities. Nonlinear normal modes are employed to understanding how the nonlinearities may influence the primary phonation. The results show that the method could reflect the most important vibratory patterns. Additionally, the nonlinear analysis has shown energy-frequency dependence and a complex dynamics with structural nonlinearities.

No contexto da fonação, as pregas vocais são responsáveis pela produção primária do sinal de voz na laringe. Tão importantes quanto complexas, elas exibem uma dinâmica rica e uma série de comportamentos que determinam as características fundamentais do sinal emitido pela boca. Com o intuito de avaliar suas propriedades sob a luz dos métodos da dinâmica não-linear de sistemas, este trabalho tem por objetivo principal desenvolver uma técnica de extração e simulação de sua dinâmica. Para isto são empregados modelos determinísticos das pregas vocais e imagens de videolaringoscopia de alta velocidade. Estes elementos são usados para reproduzir a dinâmica e cinemática das pregas de indivíduos saudáveis através de um procedimento de otimização dos parâmetros mecânicos destes modelos de pregas. Feito isto, os modelos são usados para avaliar o comportamento não-linear das pregas vocais através de uma análise modal não-linear cujo propósito foi descrever como as não-linearidades estruturais das pregas podem influenciar na voz. Os resultados mostram a eficiência de um método capaz de reproduzir a maior parte dos padrões de vibração das pregas e uma relação de dependência entre sua energia mecânica e sua frequência fundamental de vibração, que é fruto das não-linearidades estruturais das pregas vocais. / As far as phonation is concern, the vocal folds produce the primary voice signal within larynx. As important as complex, they exhibit a rich dynamics capable to influence the fundamental voice features. In light of nonlinear dynamical methods, this work has as purpose to develop a method to extract and simulate the vocal folds dynamics. Deterministic vocal folds models and high-speed videolaryngoscopy are used to reproduce the dynamics of healthy subjects and evaluate their nonlinear behavior and some phenomena raised by the nonlinearities. Nonlinear normal modes are employed to understanding how the nonlinearities may influence the primary phonation. The results show that the method could reflect the most important vibratory patterns. Additionally, the nonlinear analysis has shown energy-frequency dependence and a complex dynamics with structural nonlinearities.

INTRODUÇÃO: O grande desafio para o laringologista perante o quadro de paralisia unilateral do músculo cricotireoideo é fazer o diagnóstico do comprometimento desse músculo por meio da laringoscopia. Não existe consenso na literatura em relação aos achados vibratórios que possam servir como comprovação dessa condição. O objetivo desse estudo foi avaliar os efeitos da contração do músculo cricotireoideo sobre a vibração das pregas vocais usando a videolaringoscopia de alta velocidade e tentar encontrar um ou mais achados que pudessem ser utilizados como um sinal indicativo de paralisia unilateral do músculo cricotireoideo. MÉTODOS: Realizou-se um estudo experimental com 20 laringes excisadas de cadáveres humanos adultos masculinos. A vibração cordal foi produzida artificialmente com a passagem de ar comprimido através das pregas vocais. Cada laringe foi avaliada em três situações: contração bilateral do músculo cricotireoideo, contração unilateral do músculo cricotireoideo e ausência de contração de ambos os músculos cricotireoideos. Os seguintes parâmetros foram avaliados pela videolaringoscopia de alta velocidade: frequência fundamental, periodicidade da vibração, amplitude de vibração das pregas vocais e simetria de fase entre as pregas vocais. RESULTADOS: A diminuição da tensão longitudinal das pregas vocais decorrente da ausência de contração do músculo cricotireoideo unilateral e bilateral mostrou não alterar a periodicidade da vibração e a presença de assimetria de fase entre as pregas vocais, mas houve uma diminuição significativa da frequência fundamental (p < 0,001). Observou-se, também, um aumento da amplitude de vibração das pregas vocais direita e esquerda (p < 0,05), que apresentaram comportamento semelhante para esse parâmetro nas diferentes situações estudadas. CONCLUSÕES: A diminuição da tensão nas pregas vocais, causada pela ausência da contração do músculo cricotireoideo unilateral e bilateral, resultou em uma diminuição da frequência fundamental e em um aumento da amplitude de vibração. A presença de periodicidade em todas as laringes sugere que a videoestrobolaringoscopia, método usado mais frequentemente na prática clínica, possa ser um método adequado para avaliar pacientes com paralisia unilateral do músculo cricotireoideo. E, finalmente, os resultados sugerem que a paralisia unilateral do músculo cricotireoideo parece não ser capaz de produzir uma assimetria de tensão suficiente para induzir diferença de amplitude de vibração e de fase entre as pregas vocais e, desta forma, esses parâmetros não podem ser utilizados para fazer uma diferenciação segura dessa condição / INTRODUCTION: The great challenge for laryngologists dealing with cricothyroid muscle paralysis is to diagnose impaired cricothyroid muscle function using laryngoscopy. There is no consensus in the literature on the vibratory findings that can be used as a proof of this condition. The aim of this study was to evaluate the effects that cricothyroid muscle contraction has on vocal fold vibration as evaluated with high-speed videoendoscopy, and to identify one or more aspects of the vocal fold vibration that could be used as an indicator of unilateral cricothyroid muscle paralysis. METHODS: We conducted an experimental study on 20 larynges excised from adult male human cadavers. Laryngeal vibration was artificially produced by compressed air passing through the vocal folds. Each larynx was assessed in three situations: bilateral cricothyroid muscle contraction, unilateral cricothyroid muscle contraction and no contraction of either cricothyroid muscle. The following parameters were evaluated by high-speed videoendoscopy: fundamental frequency, periodicity, amplitude of vocal fold vibration, and phase symmetry between the vocal folds. RESULTS: Although neither unilateral nor bilateral cricothyroid muscle contraction altered periodicity of vibration or the occurrence of phase asymmetry, there was a significant decrease in fundamental frequency in parallel with the decreasing longitudinal tension (p < 0,001). We also found an increase in vibration amplitude of right and left vocal folds (p < 0,05), which were similar in terms of their behavior for this parameter in the various situations studied. CONCLUSION: Decreased vocal fold tension caused by the absence of unilateral or bilateral contraction of cricothyroid muscle resulted in a decrease in fundamental frequency and an increase in vibration amplitude. The presence of regular periodicity in all studied larynges suggests that videostroboscopy, method most frequently used in clinical practice, may be a suitable method for the evaluation of patients with unilateral cricothyroid muscle paralysis. We also believe that unilateral cricothyroid muscle paralysis is not capable of causing a sufficient degree of tension asymmetry to induce phase asymmetry or a difference in vibration amplitude between the vocal folds and, hence, these parameters can\'t be used as reliable indicators of this condition