2758 IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 58, NO. 10, OCTOBER 2011 Optical 3-D Metric Measurements of Local Vocal Fold Deformation Characteristics in an In Vitro Setup Bj¨ orn H ¨ uttner, Alexander Sutor, Member, IEEE, Georg Luegmair, Stefan J. Rupitsch, Member, IEEE, Reinhard Lerch, Senior Member, IEEE, and Michael D ¨ ollinger* Abstract—Understanding vocal fold dynamics presents an es- sential part in treating voice disorders as it is the prerequisite to appropriate medical therapy. Various physical and numerical mod- els exist for simulation purposes, all relying on simplified material parameters. To improve current approaches, data of realistic tis- sue behavior, i.e., in natural surroundings, have to be considered in model development. An in vitro setup was proposed for tensile tests combined with an optical method for precise, local and met- rical 3-D measurements of distinctive surface points. Compared to previous 3-D reconstruction methods, the accuracy was improved tenfold. Vertically applied forces versus resulting deformation were measured for ten porcine vocal folds. Deformation characteristics of mucosa and the two-layer structure of mucosa and muscle (MM) were investigated at three distinctive locations along the vocal fold edge. The spring rates were represented by an exponential func- tion. For equal deflections, an increasing spring rate from posterior to anterior for MM was measured. For solely mucosa, the spring rate decreased from the posterior to the middle and subsequently increased again. The MM-layer presented a stiffer deformation behavior than mucosa. For deformations higher than 1.5 mm, the spring rates for MM were more than twice as high as for mucosa. The investigations display the importance of considering both mul- tilayers and local differences for the improvement of vocal fold models. Index Terms—Image processing, optical measurement, stereo triangulation, tensile test, vocal fold deformation characteristics. I. INTRODUCTION S PEECH is the most important mean of expression in human communication [1]. Voice disorders, for example dyspho- nias, minimize the quality and intelligibility of speech. Thus, they heavily affect social integration and, therefore, quality of life [2]. Manuscript received August 12, 2010; revised November 11, 2010, February 8, 2011; accepted March 7, 2011. Date of publication March 22, 2011; date of current version September 21, 2011. Asterisk indicates corresponding author. B. H¨ uttner and G. Luegmair are with the Department of Phoniatrics and Pediatric Audiology, Medical School, University Hospital Erlangen, Erlangen 91054, Germany (e-mail: bjoern.huettner@uk-erlangen.de; georg. luegmair@uk-erlangen.de). A. Sutor, S. J. Rupitsch, and R. Lerch are with the Chair of Sensor Technologies, Friedrich-Alexander-University Erlangen-Nuremberg, Erlangen 91054, Germany (e-mail: Alexander.Sutor@lse.eei.uni-erlangen.de; Stefan. Rupitsch@lse.eei.uni-erlangen.de Reinhard.Lerch@lse.eei.uni-erlangen.de). *M. D¨ ollinger is with the Department of Phoniatrics and Pediatric Audiol- ogy, Medical School, University Hospital Erlangen, Erlangen 91054, Germany (e-mail: michael.doellinger@uk-erlangen.de). Digital Object Identifier 10.1109/TBME.2011.2130525 The primary voice signal originates in the larynx by two op- posing oscillating vocal folds driven by an airstream from the lungs. Commonly, a healthy voice is characterized by symmet- ric and periodic oscillations of the vocal folds [1] as well as a phase of (almost) entire glottis closure [3]. Dysphonias affect this behavior and yield a disturbed acoustical signal. They result from visible morphological alterations, for example polyps, and are easily detectable by endoscopic inspection of the larynx [4]. However, so called functional dysphonias exist, which decrease voice quality without exhibiting a visual morphological tissue alteration [3], [5]. These diseases display their clinical picture only during phonation in nonperiodic or asymmetric oscilla- tions. The actual reasons for that behavior are not yet known. In order to be able to understand healthy and pathological tissue vibrations, the interactions of air flow, structural dynamics, and acoustics have to be explored. One approach is the identifica- tion of the interrelations between oscillation patterns and tissue properties. Local information during vocal fold oscillations has been recorded by Berry et al. [6] and D ¨ ollinger et al. [7]. Qin et al. [8] applied a combination of high-speed and electroglottography (EGG) recordings to quantify vibration parameters of vocal folds. The 3-D movements of the vocal fold surface were exam- ined and quantified by Luegmair et al. [9]. To obtain more information on vocal fold tissue and its ma- terial properties, several experimental approaches have been suggested. Rheometers are used to measure the mechanical response of tissue to applied torsion and, thus, to identify shear elasticity and viscoelastic properties [10]–[13]. A lin- ear skin rheometer (LSR) is applied by Chetri et al. [14], Goodyer et al. [15], and Hess et al. [16] to deliver local information about stiffness, shear modulus, and viscosity by analyzing stress/strain data. Standing waves are induced in tissue and measured by utilizing a bioreactor [17]. Apply- ing an equation of motion with appropriate boundary condi- tions, Young’s modulus is calculated. However, the first and last of these mentioned methods deliver global mean values of tissue material parameters. Thus, the effects of local inho- mogeneities, as caused by polyps, are neglected. The experi- mental setup proposed by Hess et al. [16] suffers from influ- encing the material properties due to contact. A noninvasive method by means of air pulse stimulation of scarred rabbit vo- cal folds was presented by Herteg˚ ard et al. [18]. The effects of scarred vocal fold tissue, and, thus, locally varied tissue properties, to the acoustics and the subglottal pressure were also examined by Murugappan et al. on a canine larynx model [19]. Global material parameters and Young’s modulus were also 0018-9294/$26.00 © 2011 IEEE