<div>Vocal fold scarring is the fibrotic manifestation of most common pathological voice disorders. Voice disorders lead to direct healthcare costs of over $200 million annually and significantly reduce quality of life for patients. Despite advances in understanding the pathophysiology of vocal fold scarring, effective treatments for scarring and fibrosis remain elusive. The wound-healing cascade associated with vocal fold injury involves complex signaling interactions between cells and their extracellular matrix (ECM), which remain largely unexplored due to the lack of a physiologically relevant preclinical model to study them. Traditional preclinical models do not capture the complex 3D microenvironment of the vocal folds, and the use of stem cells or fibroblasts alone in models has resulted in poor reproducibility and predictability of in vitro models. Toward this end, this work describes the development of a preclinical model that strives to take into account cellular interactions between fibroblasts and epithelial cells and achieve a balance in the native vocal fold 3D environment to function as an in vitro model.</div><div><br></div><div>Since a major shortcoming of current in vitro models is the lack of a standardized epithelial fibroblast coculture, initial work focused on developing a coculture system between commercially available tracheal epithelial cells and vocal fold fibroblasts in an in vitro setting that would provide more accurate information about the disease pathophysiology and help design better targeted treatments. We designed a healthy and disease state coculture model that can be induced into a fibroplastic state to overexpress stress fibers using TGFβ1. We also demonstrated that both cell types maintained phenotype in the healthy and disease state coculture models.</div><div><br></div><div>To further transfer this model in a physiologically relevant 3D system, follow-up research characterized 3D matrices to mimic the native ECM of the vocal folds by using natural biomimetic materials found in the vocal folds such as hyaluronic acid, type I collagen, and type III collagen. We hypothesized that the ability to control the viscoelastic and structuralcharacteristics of the scaffold in combination with presenting relevant biological cues to cells will result in a better biomimetic scaffold. This research is expected to lay effective groundwork for developing a functional tissue engineered 3D coculture model that retains the reproducibility necessary to serve as a viable diagnostic and therapeutic screening platform.</div>

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / verify and correlate the acoustic vocal modifications of filter and glottal source, perceptual hearing, laryngeal, sound pressure level (SPL) and self-reported vocal sensations, after the phonation in a glass tube immersed in water (PGTIW), on adult women. Methods: observational study cross-sectional quantitative analytical, composed of 24 adult women, with ages between 18 and 40 years (average 23,04 years), without laryngeal affections (LA) and without vocal complaints, being 12 in the study group (SG) and 12 in the control group (CG). In the EG was held the collection of the vowel /a:/, measure of SPL, videolaryngostroboscopy (VLS), execution of the technique of PGTIW in three sets of the 15 repetitions and, immediately after, the collect of vowel /a:/, measure of SPL, VLS and reporting of sensations in relation to voice. In the CG, the same procedures were performed, but instead the implementation of technical, the volunteers remained in silence. Was performed vocal acoustic analysis with the programs Multi Dimension Voice Program Advanced and Real Time Spectrogram (Kay Pentax®). Judges speech therapyst performed the analysis of spectrographies, with specific protocol and hearing perceptual with the RASAT scale and judges otolaryngologists performed the analysis of the VLS. It was used the statistical tests Wilcoxon, Chi-square and Spearman. Results: On the SG: improvement of the smoothed pitch perturbation quotient (sPPQ), voice turbulence index (VTI), Shimmer percentage (Shim) and Shimmer in dB (ShdB); improvement of breathiness in RASATI; increase of the tracing darkening of the fourth formant, the definition of the first formant, the definition and the number of harmonics; increase the SPL and sensation better voice; in the VLS, the laryngeal vestibule constriction did not change significantly. Positive correlation between sensation better voice and the definition of the second and third formants and regularity the tracing; between measures of Shimmer and breathiness; between the fundamental frequency and maximum instability. Conclusion: in the group of women without laryngeal disorders and without vocal complaints studied, the technique generated a reduction of of vibration glottal aperiodicity and the noise, with consequent improvement of the glottal signal. In VLS, the laryngeal vestibule constriction did not change significantly, this suggests that the technique does not cause laryngeal hypertension. / verificar e correlacionar as modificações vocais acústicas de filtro e de fonte glótica, perceptivoauditivas, laríngeas, nível de pressão sonora (NPS) e sensações vocais autorrelatadas após a fonação em tubo de vidro imerso em água (FTVIA), em mulheres adultas. Métodos: estudo observacional transversal analítico quantitativo, composto por 24 mulheres adultas, com idades entre 18 e 40 anos (média 23,04 anos), sem afecções laríngeas (AL) e sem queixas vocais, sendo 12 do grupo de estudo (GE) e 12 do grupo de controle (GC). No GE, realizou-se a coleta da vogal /a:/, medida do NPS, videolaringoestroboscopia (VLE), execução da técnica de FTVIA em três séries de 15 repetições e, imediatamente após, a coleta da vogal /a:/, medida do NPS, VLE e relato de sensações em relação à voz. No GC, os mesmos procedimentos foram realizados, porém, ao invés da execução da técnica, as voluntárias permaneceram em silêncio. Realizou-se a análise acústica vocal com os programas Multi Dimension Voice Program Advanced e Real Time Spectrogram (Kay Pentax®). Juízas fonoaudiólogas realizaram a análise das espectrografias, com protocolo específico e perceptivoauditiva com a escala RASATI, e juízes otorrinolaringologistas realizaram a análise da VLE. Utilizaram-se os testes estatísticos Wilcoxon, Qui-quadrado e Spearman. Resultados: no GE: melhora do quociente de perturbação do pitch suavizado (sPPQ), índice de turbulência vocal (VTI), Shimmer percentual (Shim) e Shimmer em dB (ShdB); da soprosidade na RASATI; do escurecimento do traçado do quarto formante, da definição do primeiro formante, da definição e do número de harmônicos; aumento do NPS e da sensação de voz melhor; na VLE: a constrição do vestíbulo laríngeo não se alterou significativamente. Correlação positiva entre a sensação de voz melhor e a definição do segundo e terceiro formantes e regularidade do traçado; entre medidas de Shimmer e soprosidade; entre frequência fundamental máxima e instabilidade. Conclusão: no grupo de mulheres sem afecções laríngeas e sem queixas vocais estudado, a técnica gerou redução da aperiodicidade da vibração glótica e do ruído, com aumento da energia harmônica e consequente melhora do sinal glótico; maior NPS, ressonância e projeção vocal, melhorando as sensações subjetivas à produção vocal; na VLE, a não alteração da constrição do vestíbulo laríngeo sugere que a técnica não provoca hipertensão laríngea.

La parole et le chant ont une importance capitale dans la culture humaine. Cependant les phénomènes physiques de production et de contrôle de la voix sont encore mal compris, et leurs paramètres mal connus, principalement en raison de la difficulté d'y accéder in vivo. Dans le modèle source-filtre simplifié, la source sonore est produite par l'oscillation des plis vocaux à une fréquence fondamentale et ses multiples ; les résonances du conduit vocal filtrent l'enveloppe spectrale du signal pour produire des voyelles. Dans cette thèse, les propriétés de la source et du filtre sont étudiées et une expérience in vitro examine l'influence du filtre sur la source. L'influence des paramètres de contrôle aérodynamiques ou mécaniques sur la fréquence fondamentale est étudiée ex vivo en utilisant des larynx humains excisés. Quatre types de discontinuités ou d'hystérésis sont observés. En dehors de ces zones de bifurcation, la fréquence fondamentale est approximativement proportionnelle à la racine carrée de la pression sous-glottique, ce qui a des implications pour le chant et de la parole, en particulier dans les langues tonales. De plus, le flux d'air qui traverse la glotte provoque un rétrécissement du conduit aryépiglottique sous l'effet de forces de pression aérodynamique, et peut initier l'oscillation des plis ventriculaires et/ou aryépiglottiques sans contrôle musculaire. L'impédance acoustique de conduits vocaux fut mesurée in vivo sur un intervalle de9 octaves en fréquence et de 80 dB en amplitude, avec la glotte fermée puis pendant la phonation. Les fréquences, amplitudes et largeurs de bande des résonances acoustiques et des résonances mécaniques des tissus autour du conduit vocal sont estimées. Lorsque la glotte est fermée, les largeurs de bande et les pertes d'énergie correspondantes dans le conduit vocal sont largement supérieures aux pertes viscothermiques d'un cylindre rigide lisse, et sont encore plus importantes pendant la phonation. En utilisant un modèle simple de conduit vocal et les mesures effectuées en inspirant, des résonances acoustiques du système sous-glottique sont également estimées. Les effets possibles de la charge aéroacoustique du filtre sur la source sont mis en évidence dans une expérience sur une maquette de plis vocaux constituée de boudins de latex remplis d'eau couplés à un tuyau rigide. La modification de la charge acoustique en aval des plis vocaux, par insertion d'une paille à l'extrémité du conduit, modifie la fréquence fondamentale de vibration des plis. Ce résultat est discuté dans le contexte des méthodes de rééducation orthophonique à la paille couramment utilisées en thérapie de la parole. / Speech and singing are of enormous importance to human culture, yet the physics that underlies the production and control of the voice is incompletely understood, and its parameters not well known, mainly due to the difficulty of accessing them in vivo. In the simplified but well-accepted source-filter model, non-linear vocal fold oscillation produces a sound source at a fundamental frequency and its multiples, the resonances of the vocal tract filter the spectral envelope of the sound to produce voice formants. In this thesis, both source and tract properties are studied experimentally and an in vitro experiment investigates how the filter can affect the source. The control of fundamental frequency by either air supply or mechanical control parameters is investigated ex vivo using excised human larynges. All else equal, and excluding the four types of discontinuity or hysteresis observed, the fundamental frequency was found to be proportional to the square root of subglottal pressure, which has implications for singing and speech production, particularly in tonal languages. Additionally, airflow through the glottis causes a narrowing of the aryepiglottic tube and can initiate ventricular and/or aryepiglottic fold oscillation without muscular control. The acoustic impedance of the vocal tract was measured in vivo over a range of 9 octaves and 80 dB dynamic range with the glottis closed and during phonation. The frequencies, magnitudes and bandwidths were measured for the acoustic and for the mechanical resonances of the surrounding tissues. The bandwidths and the energy losses in the vocal tract that cause them were found to be five-fold higher than the viscothermal losses of a dry, smooth rigid cylinder, and to increase during phonation. Using a simple vocal tract model and measurements during inhalation, the subglottal system resonances were also estimated. The possible effects of the filter on the source are demonstrated in an experiment on a water-filled latex vocal fold replica: changing the aero-acoustic load of the model tract by inserting a straw at the model lips changes the fundamental frequency. This result is discussed in the context of straw phonation used in speech therapy.

Cette thèse présente la conception, la fabrication et la commande d’un microrobot parallèle à deux Degrés de liberté (rotations θ1 et θ 2) qui est dédié pour la microchirurgie laser des cordes vocales.Ce travail a été développé dans le cadre du projet européen “Micro-Technologies and Systems for Robot-Assisted Laser Phonomicrosurgery” intitulé μRALP. D’une part, les aspects médicaux et biologiques ont permis d’extraire un cahier des charges pour le microrobot. D’autre part, l’ état de l’art a permis d’identifier et d’apporter une solution aux verrous technologiques pour satisfaire toutes les requêtes. Ce travail a proposé le développement de deux microrobots indépendants avec des structures parallèles. Bien qu’ayant le même objectif qui consiste à guider deux faisceaux lasers de manière précise vers les cordes vocales, chaque microrobot aune structure cinématique différente. Un miroir placé sur la plateforme de chacun des microrobots permet le guidage des faisceaux, en effectuant la réflexion et le changement de direction. La technique“Smart Composite Microstructures” a été utilisée pour microfabriquer les deux microrobots. En outre, les mouvements ont été réalisés grâce a à l’utilisation d’actionneurs piézoélectriques. Enfin, la précision attendue par la chirurgie des cordes vocales a été atteinte en associant la commande par asservissement visuel et la pré-compensation des non linéarités et des oscillations mal amorties / This thesis reports the design, fabrication and control of a parallel microrobot with two degrees offreedom in rotation, that is destined to perform laser microsurgery on vocal folds. This work has beendeveloped in the frame and in the context of the European project Micro-Technologies and Systemsfor Robot-Assisted Laser Phonomicrosurgery entitled μRALP.On the one hand, needs and limits of medical aspects have been abstracted to create the microrobotspecifications. On the other hand, the state-of-the-art has identified the technological stumblingblocks and has allowed to led to our proposition to satisfy the specifications. This work has envisagedthe fabrication of two independent parallel microrobotic structures. Each proposed microrobot hasa different kinematic structure, however they have the same objective that is to guide a laser beamonto the vocal folds. To this purpose, a mirror is placed on the end-effector of both microrobots inorder to reflect and guide the laser beam. Smart Composite Microstructures fabrication techniquewas used to fabricate the two microrobots. Moreover, piezoelectric cantilever actuators have beenused to generate the required rotations. Finally, the expected precision for laser microsurgery of thevocal folds was reached by coupling the feedforward compensation of nonlinearities and vibrations ofthe actuators with visual feedback control of the parallel kinematic mechanism.

Developing lifelike vocal fold models is challenging due to various associatedbiomechanical complexities. Nevertheless, the development and analysis of improved vocal foldmodels is worthwhile since they are valuable tools for gaining insight into human vocal foldvibratory, aerodynamic, and acoustic response characteristics. This thesis seeks to contribute tothe development of computational and physical vocal fold modeling in two ways. First is byintroducing a method of obtaining internal deformation fields within vibrating synthetic vocal foldmodels; second is by presenting an optimization algorithm coupled with a computational vocalfold model to optimize geometry and stiffness of a synthetic vocal fold model to achieve morerealistic vibration patterns.The method for tracking the internal deformation of self-oscillating vocal fold models isbased on MR imaging. Silicone models scaled to four times life-size to lower the flow-inducedvibration frequency were imbedded with fiducial markers in a coronal plane. Candidate markermaterials were tested using static specimens, and two materials, cupric sulfate and glass, werechosen for testing in the vibrating VF models. The vibrating models were imaged using a gatedMRI protocol wherein MRI acquisition was triggered using the subglottal pressure signal. Twodimensionalimage slices at different phases during self-oscillation were captured, and in eachphase the fiducial markers were clearly visible. The process was also demonstrated using a threedimensionalscan at two phases. The benefit of averaging to increase signal-to-noise ratio wasexplored. The results demonstrate the ability to use MRI to acquire quantitative deformation datathat could be used, for example, to validate computational models of flow-induced VF vibrationand quantify deformation fields encountered by cells in bioreactor studies.A low fidelity, two-dimensional, finite element model of VF flow-induced vibration wascoupled with a custom MATLAB-based genetic algorithm optimizer. The objective was to achievea closed quotient within the normal human physiological range. The results showed that changesin geometry and stiffness would lead to a model that exhibited the desired characteristics. Aphysical model based on optimized parameters was then fabricated and the closed quotient wastested. The physical model successfully vibrated with nonzero closed quotient as predicted by thecomputational model.

Title: Numerical modelling of vocal folds Author: Mgr. Bc. Jakub Michálek Department: Geophysics Department Supervisor: RNDr. Jakub Velímský, Ph. D., Geophysics Department Abstract: I created a 2D model of vocal fold oscillations driven by interaction with fluid flow (FSI model). I used the theory of finite strains of the structure. The flow was modelled as a viscous incompressible fluid. The numerical problem was solved by the finite element method (FEM) with the arbitrary Lagrangian- Eulerian (ALE) method. The model shows the regime before phonation without the contact of the vocal folds and the states that directly lead to the contact of the vocal folds. I compared the time dependence of the quantities measured at sensors attached to the folds. I measured the frequency characteristics and compared the results with the modal analysis. Consequently, I calculated a parametric study of the dependence of the oscillations on the inlet air velocity, the original distance of the vocal folds and the elasticity of the vocal folds. The parametric study confirms that a sufficient increase of the inlet air velocity and a sufficient constriction of the glottis lead to the phonation onset. The parametric study also shows that changing elastic parameters has a similar effect on the spectrum for both the FSI model and the modal...

Vocal fold models are valuable for studying voice production. They provide an alternative method of studying the mechanics of the voice that does not require in vivo experimentation or the use of excised human or animal tissue. In this thesis, a new method of creating vocal fold models through additive manufacturing is described. The purpose of this research was to reduce model fabrication time, to decrease the number of model failures during manufacturing, and to lay the foundation for creating models with more lifelike geometric and material properties. This research was conducted in four stages. First, a suitable silicone additive manufacturing technique using a UV-curable silicone was chosen. The technique chosen was called freeform reversible embedding (FRE) and involved embedding liquid silicone material into a gel-like medium named organogel. The UV-curable silicone's material properties were identified to confirm its utility in vocal fold model design. Second, an open-source, fused deposition modeling slicing software was selected to create g-code for the printer. Applicable software settings were tuned through qualitative printing tests to find their optimal values for use in FRE printing. Third, 3D-printed cubes were used in tensile tests to characterize the material properties of FRE-printed, silicone material. The cubes were found to be anisotropic, exhibiting different modulus values corresponding to the layer orientation of the printed material. Fourth, vocal fold models were FRE-printed in two different layer orientations and were used in phonation tests to gather data for onset pressure, vibratory frequency, amplitude, and flow rate. The printed models self-oscillated and withstood the strains induced by phonation. These tests showed that layer direction affects the phonation properties of the models, demonstrating that models with layers in the coronal plane had slightly lower frequencies and onset pressures than models with layers in the sagittal plane. The models' onset pressures were higher than what is found in human vocal folds. However, their frequencies were within a comparable range. These tests showed the effectiveness of additive manufacturing in the application of vocal fold fabrication, reducing production effort by allowing researchers to go directly from model design to fabrication in a single manufacturing step. It is anticipated that this method will be modified to incorporate printing of multiple stiffnesses of silicone to better mimic the material properties of vocal fold tissue, and that the anisotropy of 3D-printed material will be leveraged to model the anisotropy of human vocal folds. This work also has potential application areas outside of voice research.

Laryngeal cancer affects tens of thousands worldwide every year. The standard of care of surgical resection, chemotherapy, and/or radiation therapy results in significant quality of life deficits including reliance on tracheostomy tubes, loss of voice, and inability to swallow. There are no therapeutic options that restore a functional larynx so that patients can live a more normal life. Laryngeal reconstruction using tissue engineering strategies offers the potential to solve this problem. Laryngeal anatomy is complex with multiple tissue types and therefore engineering approaches require consideration of a multi-layer, interfacial tissue design. Our strategy to overcome these challenges involves the use of advanced bio fabrication techniques where type I oligomeric collagen alone or in the presence of autologous stem cells is used to custom-make the cartilage, skeletal muscle, and mucosal layer of the larynx. This doctoral research project begins by describing the development of the tissue engineered skeletal muscle with aligned collagen matrix and autologous muscle progenitor cells induced to express motor endplates. Next, using this engineered muscle plus the cartilage layer developed by a colleague; we implanted the myochondral engineered construct in a rat hemilaryngectomy model. In this study we saw host-implant integration with no inflammatory foreign body response, neo cartilage and muscle formation, and some return of laryngeal function on the reconstructed side. Next, we worked to scale-up these technologies for use in a porcine model. The pig larynx is more similar in size and function to the human larynx and allows for a full thickness defect to be created. Using confined compression, we created 4-mm thick acellular and cellular cartilage constructs, as well as a 0.5-mm thick acellular mucosal layer. A 1-cm diameter muscle layer containing autologous muscle progenitor cells was created using flow alignment and cultured to induce expression of motor endplates before implantation. Tissue constructs were subjected to mechanical property analyses as well as PCR analysis to describe the differential gene expression by component cells within muscle and cartilage constructs. Each layer was individually sutured into a pig hemilaryngectomy model. The pigs recovered well from the surgery, were eating, had no difficulty breathing, and no aspiration events. At 2 months, respiratory epithelium had completely healed over the implant and was vascularized and had areas of submucosal gland growth. The motor endplate expressing muscle implants showed new, organized muscle ingrowth while the acellular implants showed a relative paucity of new, disorganized muscle. This work represents a significant advancement in the field of laryngeal reconstruction and is a first of its kind to use scalable tissue engineering technologies designed to specifically meet each layer’s functional criteria.

Communication by voice is foundational in our society and many rely on their voices for their occupations. Voice disorders affect a significant number of individuals each year, and diagnosis and treatment improvements are therefore sought via advancements in voice research. Contained in this thesis is a description of work intended to contribute to vocal fold research by using synthetic, self-oscillating vocal fold replicas to study the impact of replica vibration on perfusion fluid flow through the replica. Methods for manufacturing vocal fold replicas containing imbedded channels allowing for fluid perfusion are discussed. Experimental procedures developed for delivering perfusion fluid to the imbedded channel at a constant pressure during replica vibration are described. Methods for measuring perfusion parameters of perfusion fluid pressure, imbedded channel diameter, flow rate, and vibration parameters (subglottal pressure, frequency, amplitude, channel length, and glottal width) are detailed. Experiments performed using both stationary and vibrating vocal fold replicas are presented. Correlations between perfusion pressure and channel diameter are discussed. Vibration parameters were correlated to perfusion flow parameters and it is shown that perfusion flow rate through the channels decreased significantly with model vibration. Potential mechanisms for changes in perfusion flow resistance with vibration are discussed and analyzed. Output of a theoretical model, developed to incorporate some of the possible flow resistance mechanisms, was compared to vibrating replica experimental data.

The vocal folds are a complex self-oscillating biological system. In the current research, an equation was developed to model viscous adhesion forces that occur when the collision of the vocal folds results in the formation of a liquid bridge. The adhesion equation was validated using experimental data, and simplified to a one-dimensional approximation with an included correction factor that adjusted the predicted pressure in situations where the one-dimensional approximation was invalid. A non-oscillating vocal fold model with a modeled liquid bridge was used to study stress resulting from viscous adhesion. The vertical normal stress magnitude ranged from about 80 to 1700 Pa. This was shown to be of the same order of magnitude as the stress due to collision of the vocal folds. Also the stress resulted in large normal strains that occurred at small distances below the surface of the vocal folds consistent with lesion development. Therefore, it was determined that the viscous adhesion may be a contributor to damage of the vocal folds that leads to the development of benign lesions, such as vocal nodules. This conclusion was further validated by adding the adhesion equation in a self-oscillating vocal fold model. The influence of adhesion on the dynamics of the model was significant. The frequency of vibration was reduced by nearly 2.5% for the case of adhesion with a mucus viscosity of 0.01 Pa-s. Also adhesion induced positive tensile stress that resulted in normal strain distributions similar to those seen in the non-oscillating cases. These results also indicated that liquid mediated viscous adhesion may be a contributor to the development of benign lesions (nodules). However, further research is needed to validate these conclusions.