Nonlinearity may be central to the interpretation of procedures used to assess irregularity in voice disorders. Lack of understanding nonlinearity may help explain why many traditional clinical tests have failed to be used routinely in the clinic. Greater knowledge of bifurcation and chaos should enhance our understanding and assessment of laryngeal pathologies, and our ability to recommend appropriate treatment. This proposal focuses on applying chaos theory to the study of normal and disordered phonation. The study has three interrelated parts. In part I, research will focus on the chaotic dynamics of the lumped-parameter model of vocal folds and finite element analysis (FEA). Using nonlinear dynamic methods, we will quantitatively describe irregular dynamics of vocal folds, such as those in cases of asymmetric vocal folds and vocal polyps. FEA will be coupled with the nonlinear pressure-flow relation in the glottis. The effects of subglottal pressure, vocal fold tension, stiffness, mass, and vocal prephonatory parameters will be examined using a bifurcation diagram. Chaotic synchronization will be used to extract biomechanical parameters in the chaotic dynamic model of vocal folds. In part II, asymmetry, vocal polyps, and dehydration will be modeled with excised larynx and evaluated with nonlinear dynamic methods. The irregular dynamics of above range phonation (ARP) will be quantitatively addressed, and phonation instability pressure (PI P) will be measured. The effects of vocal fold tension, mass, stiffness, asymmetric vocal folds, and vocal polyps on PIP and ARP will be investigated to examine the predicted results obtained in the computer models of part I. Chaotic flesh movement and glottal area will be directly recorded on high-speed video. Irregular spatio-temporal vibratory patterns will be studied. In part III, using recordings of the acoustic signal, EGG, and PGG, we will analyze nonlinear dynamic characteristics of data obtained from patients with vocal nodules and polyps, vocal fold paralysis, laryngeal carcinoma, and Parkinson's disease. The patients' voices will be compared pre and post treatment. The sensitivity and specificity of nonlinear dynamic parameters in distinguishing normal from pathologic voices will be assessed based on the received operating characteristic (ROC) analysis. ? ?
Showing the most recent 10 out of 53 publications
