Master of Arts (MA)
Communication Sciences and Disorders
High-speed digital imaging (HSDI) has recently become clinically available for the direct observation of vocal fold movement in the last 20 years. However, before it can become routinely used in the clinical setting, a universal means of objectively analyzing and interpreting the HSDI data must be established. In this study, preliminary data was gathered for five parameters used to objectively analyze vocal fold vibratory patterns observed with HSDI. The parameters investigated were established by Ikuma, Kunduk, and McWhorter (2012a) and were previously studied with a small sample (N=8) comparing pre and post-phonosurgical removal of benign lesions. The five parameters included fundamental frequency standard deviation (F0SD), harmonics-to-noise ratio (HNR) mean, open quotient (OQ) mean, speed index (SI) mean, and relative glottal gap (RGG) mean. The current study aimed to statistically and visually analyze measurements of the five objective parameters for differences between pathology groups with different etiologies. High-speed videos (N=50) were divided into five groups based on one of the following medical diagnoses: normal voice, vocal fold nodules, polyps, true vocal fold motion impairment (TVFMI), and adductor spasmodic dysphonia (ADSD). Statistical analysis showed that HNR mean differentiated normal voices from ADSD voices and that F0 mean differentiated ADSD voices from all groups except vocal fold nodules (p < 0.005). Visual analysis revealed a strong trend for RGG mean to differentiate vocal fold nodules from all other groups. Less prominent visual trends for OQ mean and SI mean were also noted.
Document Availability at the Time of Submission
Release the entire work immediately for access worldwide.
Perkins, Aimee Elaine, "Differentiation of voice disorders using objective parameters from harmonic waveform modeling in high-speed digital imaging" (2013). LSU Master's Theses. 2228.