About Yuling Yan, PhD
Fig. 2: Conventional endoscopic view (left) and the virtual
endoscopic view (reconstructed from volumetric CT data)
of the vocal folds of a patient suffering from laryngeal tumor.
Overview of Research Program and Objectives
The primary objective of my research program is to understand the mechanism of phonation for normal and for pathological voice conditions. We employ an interdisciplinary approach to these studies that borrows and integrates concepts and methodologies from bioengineering, biophysics, mathematical modeling and physiology.
1. Functional Analysis and Modeling of Phonation in Normal and Diseased States
Vibration of the vocal folds is an essential yet poorly understood event in human voice production. An important aspect of our research program is to characterize the dynamic behavior of the vocal folds during phonation - the ultimate goal for these studies is to understand the mechanism of phonation in terms of the generation and interaction of sound waves in the vocal system; these studies will lead to the development of quantitative biomechanical models of vocal fold dynamics and acoustic interactions in the vocal tract for the detection, diagnosis and assessment of treatments for specific voice disorders.
2. Quantitative analysis of vocal fold dynamics using High Speed Digital Imaging (HSDI)
HSDI with simultaneously acquired acoustic recordings are being used to characterize vocal fold dynamics. We have developed new methods and software platforms to generate comprehensive, functional analysis of vocal fold vibrations from HSDI and acoustic recordings.
Figure 1: Top) A montage of 10 image frames from an HSDI recording of a normal subject while producing a sustained vowel phonation; Bottom) Spatially resolved vocal fold vibration representing diplophonic voice and Nyquist pattern showing the bifurcation (transition from a normophonic (red) to a diplophonic phase (black).
For example, our analytical platform that integrates automatic image segmentation of the vocal folds and detection of vocal fold edge (Fig. 1, top) with the generation of glottal waveforms that include the glottal area waveform, glottal width function and displacements of the left-right vocal fold edges at specific anterior-medial-posterior locations (Fig. 1, bottom left). The approach also integrates our ‘Nyquist’ plot based waveform analysis (Yan et al., 2005. J. Voice), which provides not only an at-a-glance assessment of the vibratory properties of the vocal fold (Fig. 1, bottom right) but a comprehensive and quantitative, high-resolution description of the vibratory properties of the vocal fold for diagnosing specific voice disorders and assessment of therapies. A related analysis has been described for acoustic signals (Yan et al, 2006. J. Voice). These studies are advancing towards a better understanding of vocal mechanism and are currently under clinical evaluation for the differential diagnosis of voice disorders associated with neurological disease and the aging process. A near-term research goal is to develop a large, comprehensive and comparative database of dynamic characteristics of vocal folds derived from our image and acoustic-based analyses that will be used to correlate changes in the vibratory properties of the vocal fold with specific voice condition and pathologies. The database can be used for on-line clinical diagnoses and for training voice researchers, clinicians and medical students.
3. Virtual Laryngoscopy
Endoscopy is a routine, minimally invasive imaging technique for evaluating the three-dimensional (3D) features and properties of the inner surface of the lumen. We are extending principles and methods from Virtual Endoscopy (VE), to develop a virtual laryngoscope (VL) for non-invasive exploration of the laryngeal system for specific applications in medicine, medical education and surgery. The VL uses software to assemble data from diverse imaging techniques (e.g., CT and MRI) to reconstruct the internal anatomy in 3D. Computer rendering provides a continuous luminal view, within which one can navigate along inner surfaces, just as in conventional laryngoscope. In addition, the VL can display a perspective global view in 3D and a view of the related CT and MRI slices for informative and interactive examination for diagnosis and treatment of disease. The VLoffers several benefits over optical endoscopes that include both internal unconventional views and external anatomical views of the airway and the sub-glottal cavity in patients with infection, inflammation and neoplasia of the lumen. VL maybe especially useful for patients with stenosis, congenital defects or those unfit for general anesthesia. We are exploiting the advantages of the VL for applications in surgical examinations of sub-glottal cavity and diagnoses of laryngeal and airway diseases.
4. Functional and metabolic imaging/spectroscopy of the larynx
We are exploring specific molecular, cellular and physiological imaging techniques as part of our integrative approach to study the human vocal system. In one project we aim to use a modified endoscope to generate optical-derived information on oxygenation and de-oxygenation from near infra-red measurements of hemoglobin within exposed capillaries and to map the contractile properties of laryngeal muscle during phonation. We anticipate that changes in the metabolic and function of laryngeal tissue when used in combination with HSDI/Nyquist analysis will greatly improve the diagnosis and treatment of voice disorders.
