Attenuation Coefficients, Pixel Intensity, and Texture Analysis as Quantitative Parameters for Analyzing Optical Coherence Tomography Images of Vocal Fold Tissue

Abstract

Optical coherence tomography is a promising technology to noninvasively assess vocal fold microanatomy. However, many studies using optical coherence tomography to evaluate vocal fold tissue are largely qualitative in nature and lack quantitative measures that might be useful in clinical decision-making. The goal of this thesis was to identify quantitative metrics that can be extracted from optical coherence tomography vocal fold imaging and applied to the surgical subspecialty of Laryngology. The first goal of this study was to determine if optical coherence tomography was sensitive enough to distinguish subtle differences between vocal folds based on vocal fold content using attenuation coefficients. The second goal of this study was to determine how collagen content affects optical coherence tomography vocal fold imaging and to identify any parameters that could reflect this effect. The work in this thesis demonstrates that quantitative parameters extracted from optical coherence tomography images of vocal fold tissue can aid in understanding normal structure and disease. Optical coherence tomography can likely distinguish young from old vocal folds by studying attenuation coefficients of optical signal. This may be important for the timing of phonosurgical procedures in pediatric patients. Optical coherence tomography can also distinguish varying levels of collagen content which may become important for monitoring and treating patients with vocal fold scarring. Combined, this thesis shows that careful image acquisition and analysis can increase the clinical and research utility of optical coherence tomography in the field of Laryngology.