Faculty Bibliography

OBJECTIVES/HYPOTHESIS/OBJECTIVE:The objectives of this study were to (1) identify optimal clusters of 15 standard acoustic and aerodynamic voice metrics recommended by the American Speech-Language-Hearing Association (ASHA) to improve characterization of patients with primary muscle tension dysphonia (pMTD) and (2) identify combinations of these 15 metrics that could differentiate pMTD from other types of voice disorders. STUDY DESIGN/METHODS:Retrospective multiparametric METHODS: Random forest modeling, independent t-tests, logistic regression, and affinity propagation clustering were implemented on a retrospective dataset of 15 acoustic and aerodynamic metrics. RESULTS:Ten percent of patients seen at the New York University (NYU) Voice Center over two years met the study criteria for pMTD (92 out of 983 patients), with 65 patients with pMTD and 701 of non-pMTD patients with complete data across all 15 acoustic and aerodynamic voice metrics. PCA plots and affinity propagation clustering demonstrated substantial overlap between the two groups on these parameters. The highest ranked parameters by level of importance with random forest models-(1) mean airflow during voicing (L/sec), (2) mean SPL during voicing (dB), (3) mean peak air pressure (cmH2O), (4) highest F0 (Hz), and (5) CPP mean vowel (dB)-accounted for only 65% of variance. T-tests showed three of these parameters-(1) CPP mean vowel (dB), (2) highest F0 (Hz), and (3) mean peak air pressure (cmH2O)-were statistically significant; however, the log2-fold change for each parameter was minimal. CONCLUSION/CONCLUSIONS:Computational models and multivariate statistical testing on 15 acoustic and aerodynamic voice metrics were unable to adequately characterize pMTD and determine differences between the two groups (pMTD and non-pMTD). Further validation of these metrics is needed with voice elicitation tasks that target physiological challenges to the vocal system from baseline vocal acoustic and aerodynamic ouput. Future work should also place greater focus on validating metrics of physiological correlates (eg, neuromuscular processes, laryngeal-respiratory kinematics) across the vocal subsystems over traditional vocal output measures (eg, acoustics, aerodynamics) for patients with pMTD. LEVEL OF EVIDENCE/METHODS:II.

INTRODUCTION/BACKGROUND:The COVID-19 pandemic necessitated a rapid restructuring of the clinical management of voice and upper airway disorders by speech-language pathologists (SLPs). As in-person therapy sessions were suspended, voice-specialized SLPs across healthcare settings shifted to online teletherapy. In this survey study, we queried voice therapists on their experiences with and opinions regarding the adoption of teletherapy into routine clinical practice. METHODS:Voice-specialized SLPs were recruited nationwide to complete an online survey which included questions about the usability of software and hardware, patient management, the effectiveness of therapy, overall satisfaction, and suggestions for improvement. RESULTS:48 participants completed the survey. The majority of respondents reported frequent technical difficulties and poor access to or understanding of appropriate equipment. Overall, participants endorsed better patient access, attendance, and compliance, as well as increased scheduling flexibility. While 95% of the respondents stated they would recommend teletherapy to another SLP, only 20% supported a shift to exclusively virtual sessions. Forty percent of respondents endorsed a hybrid model consisting of initial in-person sessions followed by virtual ones. DISCUSSION/CONCLUSIONS:Incorporating teletherapy into clinical voice practice has, for the most part, followed Carl May's normalization process theory framework, in that clinicians have invested understanding, training, time and effort, and appraisal into its implementation. However, the unusually rapid pace of change necessitated by the pandemic has presented its own set of challenges. Given the inherent conveniences of virtual therapy, the online modality is likely here to stay. It is critical that we understand the facilitators and barriers to its successful adoption.

BACKGROUND/OBJECTIVES/OBJECTIVE:Base of tongue (BOT) dysfunction is common following oropharyngeal concurrent chemoradiation therapy (CCRT). We present a clinically relevant animal model quantifying the effects of CCRT on tongue strength and elasticity over time. METHODS:Fifty-three male and 53 female Sprague-Dawley rats were randomized to control or experimental groups. Experimental animals received cisplatin, 5-fluorouracil, and 5 fractions of 7 Gy directed to the BOT. Controls received no intervention. At 2 weeks, 5 months, or 10 months after CCRT, animals underwent non-survival surgery to measure twitch and tetanic tongue strength, which were analyzed using multivariate linear mixed effects models. Tongue displacement, a surrogate for tongue elasticity, was also determined via stress-strain testing and analyzed via a multivariate linear mixed effects model. RESULTS:Reporting the combined results of both sexes, the estimated experimental group mean peak twitch forces became more divergent over time compared to controls, being 8.3% lower than controls at 2 weeks post-CCRT, 15.7% lower at 5 months, and 31.6% lower at 10 months. Estimated experimental group mean peak tetanic forces followed a similar course and were 2.9% lower than controls at 2 weeks post CCRT, 20.7% lower at 5 months, and 27.0% lower at 10 months. Stress-strain testing did not find CCRT to have a significant effect on tongue displacement across experimental timepoints. CONCLUSIONS:This study demonstrates an increasing difference in tongue strength over time between controls and animals exposed to CCRT. Tongue elasticity was not significantly affected by CCRT, suggesting that changes in strength may not be caused by fibrosis. LEVEL OF EVIDENCE/METHODS:NA Laryngoscope, 2022.

OBJECTIVES/UNASSIGNED:Early in the COVID-19 pandemic, outpatient visits were adapted for the virtual setting, forcing laryngologists to presume certain diagnoses without the aid of laryngoscopy, solely based on history and the limited physical exam available via video visit. This study aims to examine the accuracy of presumptive diagnoses made via telemedicine, compared to subsequent in-person follow up, where endoscopic examination could confirm or refute suspected diagnoses. METHODS/UNASSIGNED:A retrospective chart review was conducted of 38 patients evaluated for voice-related issues at NYU Langone Health and the University of California-San Francisco. Presumptive diagnoses at the initial telemedicine encounter were noted, along with diagnostic cues used for clinical reasoning and recommended treatment plans. These presumptive diagnoses were compared to diagnoses and plans established following laryngoscopy at follow-up in-person visits. RESULTS/UNASSIGNED:After laryngoscopy at the first in-person visit, 38% of presumptive diagnoses changed, as did 37% of treatment plans. The accuracy varied among conditions. Muscle tension dysphonia and Reinke's edema were accurately diagnosed without laryngoscopy, but other conditions, including vocal fold paralysis and subglottic stenosis, were not initially suspected, relying on laryngoscopy for diagnosis. CONCLUSIONS/UNASSIGNED:While some laryngologic conditions may be reasonably identified without in-person examination, laryngoscopy remains central to definitive diagnosis and treatment. Telemedicine can increase access to care, but it may provide more utility as a screening tool, triaging which patients should present more urgently for in-person laryngoscopy. LEVEL OF EVIDENCE/UNASSIGNED:4.

OBJECTIVES/OBJECTIVE:To examine the effects of short-term and long-term engagement with structured choral singing on vocal function and quality of life outcomes in older adults. METHODS:Two groups of older adult singers over 55 years, one with fewer than 4 semesters and one with 4 or more semesters singing in a chorale, were assessed at 3 time points: baseline, after 1 semester of singing, and either after 1 semester of rest or after 1 semester of rest and 1 semester more of singing. Acoustic and aerodynamic measures, voice-related quality of life ratings, and measures of singing accuracy were obtained. Percent change between time points were calculated to determine three outcomes: improvement, lack of change, or worsening of measures across time. RESULTS:Long-term average spectrum (LTAS), difference in first and second harmonics and estimated subglottic pressure were significantly more likely to improve after a semester of singing with less experience singers, and LTAS continued to improve after a semester of rest. Flow was significantly more likely to improve with more singing experience after a semester of singing. Aerodynamic variables consistently changed in more experienced singers and improvement was maintained over the three visits. No significant changes occurred over time for singing accuracy for any singer type. Self-perception of singing voice continued to improve with more singing experience. CONCLUSIONS:This study demonstrated that for older adults in good health, regular singing provided a mechanism for maintaining speaking voice over time.

Vocal fatigue is a measurable form of performance fatigue resulting from overuse of the voice and is characterized by negative vocal adaptation. Vocal dose refers to cumulative exposure of the vocal fold tissue to vibration. Professionals with high vocal demands, such as singers and teachers, are especially prone to vocal fatigue. Failure to adjust habits can lead to compensatory lapses in vocal technique and an increased risk of vocal fold injury. Quantifying and recording vocal dose to inform individuals about potential overuse is an important step toward mitigating vocal fatigue. Previous work establishes vocal dosimetry methods, that is, processes to quantify vocal fold vibration dose but with bulky, wired devices that are not amenable to continuous use during natural daily activities; these previously reported systems also provide limited mechanisms for real-time user feedback. This study introduces a soft, wireless, skin-conformal technology that gently mounts on the upper chest to capture vibratory responses associated with vocalization in a manner that is immune to ambient noises. Pairing with a separate, wirelessly linked device supports haptic feedback to the user based on quantitative thresholds in vocal usage. A machine learning-based approach enables precise vocal dosimetry from the recorded data, to support personalized, real-time quantitation and feedback. These systems have strong potential to guide healthy behaviors in vocal use.

OBJECTIVE:Objective evaluation of physiological responses using non-invasive methods for the assessment of vocal performance and voice disorders has attracted great interest. This paper, for the first time, aims to implement and evaluate perilaryngeal-cranial functional muscle networks. The study investigates the variations in topographical characteristics of the network and the corresponding ability to differentiate vocal tasks. METHOD/METHODS:Twelve surface electromyography (sEMG) signals were collected bilaterally from six perilaryngeal and cranial muscles. Data were collected from eight subjects (four females) without a known history of voice disorders. The proposed muscle network is composed of pairwise coherence between sEMG recordings. The network metrics include (a) network degree and (b) weighted clustering coefficient (WCC). RESULTS:|=0.12) in differentiating the vocal tasks. CONCLUSION/CONCLUSIONS:Perilaryngeal-cranial functional muscle network was proposed in this paper. The study showed that the functional muscle network could robustly differentiate the vocal tasks while the classic assessment of muscle activation fails to differentiate. SIGNIFICANCE/CONCLUSIONS:For the first time, we demonstrate the power of a perilaryngeal-cranial muscle network as a neurophysiological window to vocal performance. In addition, the study also discovers tasks with the highest network involvement, which may be utilized in the future to monitor voice disorders and rehabilitation.

Pulmonary drug delivery aims to deliver particles deep into the lungs, bypassing the mouth-throat airway geometry. However, micron particles under high flow rates are susceptible to inertial impaction on anatomical sites that serve as a defense system to filter and prevent foreign particles from entering the lungs. The aim of this study was to understand particle aerodynamics and its possible deposition in the mouth-throat airway that inhibits pulmonary drug delivery. In this study, we present an analysis of the aerodynamics of inhaled particles inside a patient-specific mouth-throat model generated from MRI scans. Computational Fluid Dynamics with a Discrete Phase Model for tracking particles was used to characterize the airflow patterns for a constant inhalation flow rate of 30 L/min. Monodisperse particles with diameters of 7 μm to 26 μm were introduced to the domain within a 3 cm-diameter sphere in front of the oral cavity. The main outcomes of this study showed that the time taken for particle deposition to occur was 0.5 s; a narrow stream of particles (medially and superiorly) were transported by the flow field; larger particles > 20 μm deposited onto the oropharnyx, while smaller particles < 12 μm were more disperse throughout the oral cavity and navigated the curved geometry and laryngeal jet to escape through the tracheal outlet. It was concluded that at a flow rate of 30 L/min the particle diameters depositing on the larynx and trachea in this specific patient model are likely to be in the range of 7 μm to 16 μm. Particles larger than 16 μm primarily deposited on the oropharynx.

Endoscopic high-speed video (HSV) systems for visualization and assessment of vocal fold dynamics in the larynx are diverse and technically advancing. To consider resulting "concepts shifts" for neural network (NN)-based image processing, re-training of already trained and used NNs is necessary to allow for sufficiently accurate image processing for new recording modalities. We propose and discuss several re-training approaches for convolutional neural networks (CNN) being used for HSV image segmentation. Our baseline CNN was trained on the BAGLS data set (58,750 images). The new BAGLS-RT data set consists of additional 21,050 images from previously unused HSV systems, light sources, and different spatial resolutions. Results showed that increasing data diversity by means of preprocessing already improves the segmentation accuracy (mIoU + 6.35%). Subsequent re-training further increases segmentation performance (mIoU + 2.81%). For re-training, finetuning with dynamic knowledge distillation showed the most promising results. Data variety for training and additional re-training is a helpful tool to boost HSV image segmentation quality. However, when performing re-training, the phenomenon of catastrophic forgetting should be kept in mind, i.e., adaption to new data while forgetting already learned knowledge.

OBJECTIVE:Past literature indicates that vibrato measurements of singers objectively changed (i.e., vibrato rate decreased and vibrato extent increased) from 1900 to the present day; however, historical audio recording technology may distort acoustic measurements of the voice output signal, including vibrato. As such, the listener's perception of historical singing may be influenced by the limitations of historical technology. This study attempts to show how the wax cylinder phonograph system-the oldest form of mass-produced audio recording technology-alters the recorded voice output signal of modern-day singers and, thus, provides an objective lens through which to study the effect(s) of historical audio recording technology on vibrato measurements. METHODS:for female singers, three times into a flat-response omnidirectional microphone and onto an Edison Home Phonograph simultaneously. The middle 1-3 seconds (6-10 vibrato cycles) of each sample was analyzed for vibrato rate, vibrato extent, jitter (ddp), shimmer (dda), and fundamental frequency for each recording condition (wax cylinder phonograph or microphone). Steady-state and frequency-modulating sinewave test signals were also recorded under the multiple recording conditions. RESULTS:Results indicated no significant effect of recording condition on vibrato rate (mean [standard deviation], cylinder: 5.3 Hz [0.5], microphone: 5.3 Hz [0.5]) and no significant difference was found for mean fundamental frequency (cylinder: 389 Hz [137], microphone: 390 Hz [137]). A significant main effect of recording condition was found for vibrato extent (cylinder: ±103 cents [30], microphone: ±100 cents [31]). Additionally, mean jitter (ddp) (cylinder: 1.22% [1.09], microphone: 0.24% [0.12]) and mean shimmer (dda) (cylinder: 9.40% [4.90], microphone: 1.92% [0.94]) were significantly higher for the cylinder recording condition, indicating more cycle-to-cycle variability in the wax cylinder recorded signal. Analysis of test signals revealed similar patterns based on recording condition. DISCUSSION/CONCLUSIONS:This study validates past scholarly inquiry about vibrato measurements as extracted from digitized wax cylinder phonograph recordings by demonstrating that measured vibrato rate remains constant during both recording conditions. In other words, vibrato rate as measured from historical recordings can be viewed as an accurate representation of the historical singer being studied. Furthermore, it suggests that the value of prior vibrato extent measurements from these acoustic recordings may be slightly overestimated from the original voice output signals produced by singers near the beginning of the 20th century (i.e., a narrow vibrato extent might have been numerically smaller on average). Increased jitter and shimmer in the wax cylinder recording conditions may be indicative of nonlinearities in the phonograph recording or playback systems.