Telehealth voice assessment by speech language pathologists during a global pandemic using principles of a primary contact model: an observational cohort study protocol

Evaluating the Diagnostic Potential of Connected Speech for Benign Laryngeal Disease Using Deep Learning Analysis

OBJECTIVES

This study aimed to evaluate the performance of artificial intelligence (AI) models using connected speech and vowel sounds in detecting benign laryngeal diseases.

STUDY DESIGN

Retrospective.

METHODS

Voice samples from 772 patients, including 502 with normal voices and 270 with vocal cord polyps, cysts, or nodules, were analyzed. We employed deep learning architectures, including convolutional neural networks (CNNs) and time series models, to process the speech data. The primary endpoint was the area under the receiver's operating characteristic curve for binary classification.

RESULTS

CNN models analyzing speech segments significantly outperformed those using vowel sounds in distinguishing patients with and without benign laryngeal diseases. The best-performing CNN model achieved areas under the receiver operating characteristic curve of 0.895 and 0.845 for speech and vowel sounds, respectively. Correlations between AI-generated disease probabilities and perceptual assessments were more pronounced in the connected-speech analyses. However, the time series models performed worse than the CNNs.

CONCLUSION

Connected speech analysis is more effective than traditional vowel sound analysis for the diagnosis of laryngeal voice disorders. This study highlights the potential of AI technologies in enhancing the diagnostic capabilities of speech, advocating further exploration, and validation in this field.

Are smartphones and low-cost external microphones comparable for measuring time-domain acoustic parameters?

PURPOSE

This study examined and compared the diagnostic accuracy and correlation levels of the acoustic parameters of the audio recordings obtained from smartphones on two operating systems and from dynamic and condenser types of external microphones.

METHOD

The study included 87 adults: 57 with voice disorder and 30 with a healthy voice. Each participant was asked to perform a sustained vowel phonation (/a/). The recordings were taken simultaneously using five microphones AKG-P220, Shure-SM58, Samson Go Mic, Apple iPhone 6, and Samsung Galaxy J7 Pro microphones in an acoustically insulated cabinet. Acoustic examinations were performed using Praat version 6.2.09. The data were examined using Pearson correlation and receiver-operating characteristic (ROC) analyses.

RESULTS

The parameters with the highest area under curve (AUC) values among all microphone recordings in the time-domain analyses were the frequency perturbation parameters. Additionally, considering the correlation coefficients obtained by synchronizing the microphones with each other and the AUC values together, the parameter with the highest correlation coefficient and diagnostic accuracy values was the jitter-local parameter.

CONCLUSION

Period-to-period perturbation parameters obtained from audio recordings made with smartphones show similar levels of diagnostic accuracy to external microphones used in clinical conditions.

Usage of Heat and Moisture Exchange Devices, Virtual Visits, Masking, and Vaccinations Among People With a Laryngectomy During COVID-19

PURPOSE

This study described the COVID-19 risk mitigation actions of people with a total laryngectomy (TL) during the pandemic.

METHOD

An online survey was completed by 215 people with a TL who lived in the United States. The survey was open from December 1, 2021, to January 15, 2022.

RESULTS

There was a significant increase in frequency of heat and moisture exchange (HME) device use during the pandemic compared with pre-COVID-19. Frequency of HME use was significantly greater for those who were vaccinated and those who had at least one clinical visit with their speech-language pathologist (SLP). The use of virtual visits increased from 9% pre-COVID-19 to 37% during the pandemic. Seventy percent of respondents were "satisfied" or "very satisfied" with virtual visits and 51% judged them "as good as in-person." Eighty percent were vaccinated for COVID-19 and 75% received a booster. One third reported that they did not wear mask over the face or over the tracheostoma. Twenty percent had tested positive for COVID-19 with 70% of these people requiring hospitalization.

CONCLUSIONS

HME use and virtual SLP visits increased during the pandemic and the vaccination rate was high among this group of respondents. Overall, there were still large percentages of people with a TL who were not using an HME, not vaccinated, and did not wear a mask. SLPs should consider reaching out directly to their TL caseload, particularly those not yet seen during the pandemic, to support uptake of COVID-19 mitigation activities specific to people with a TL as the pandemic persists.

A Human-Centered Design Approach to SOVTE Straw Phonation Instruction

OBJECTIVES

Extensive research has demonstrated that straw/tube phonation exercises produce positive effects on the voice quality of the user. The purpose of this study was to explore pedagogical practices employed by a range of voice professionals when using SOVTEs by applying human-centered design approaches to examine workflow and communication between teachers and students.

METHOD/DESIGN

The researchers conducted a series of five participatory role-playing workshop activities at a multi-disciplinary vocology conference that included speech-language pathologists, voice scientists, physicians, singing teachers, acting voice teachers, choral directors, and singers. Volunteers were asked to assume the role of teaching or learning straw phonation exercises. The other participants from each workshop were asked to provide written and verbal feedback regarding the teacher/student dynamics and what they found most important and relevant in terms of best practices for SOVTEs. Expert evaluators subsequently provided similar feedback based on video recordings of the workshops.

RESULTS

While there was variability noted in teaching techniques and results, common themes emerged in the analysis. Successful and unsuccessful student and teacher techniques were identified by the workshop participants. Through the video analysis, the research team identified themes in instructional teaching, experiential learning, and straw articulation and mechanics.

CONCLUSIONS

The benefits of straw/tube phonation exercises on voice quality have been well-established and are clear. As we continue to refine best practices in SOVTEs and, in particular, straw phonation, involvement of experts in HCD as part of multidisciplinary teams may lead to the eventual design of SOVTE tools that provide greater precision, consistency, and effectiveness both in clinical and pedagogical settings, and aid in the differentiation of SOVTE approaches within the taxonomy of vocal habilitation and rehabilitation techniques.

COVID-19 and the resurgence of telehealth in otolaryngology

The objective of this review is to examine the impact of the ongoing Coronavirus disease 2019 (COVID-19) pandemic on the use of telehealth in Otolaryngology. The use of telemedicine rose dramatically during the pandemic to meet the need for continued patient care while allowing for physical separation of providers and patients. Telemedicine has been used to evaluate patients with a variety of pathologies including dysphonia, vertigo, and anosmia. Innovative use of at-home exams, such as video-otoscopy has aided providers in overcoming challenges associated with a highly specialized physical exam. The use of telemedicine in Otolaryngology will likely remain essential in the post-pandemic era and has promising results for improving clinical efficiency.

Lightweight Deep Learning Model for Assessment of Substitution Voicing and Speech after Laryngeal Carcinoma Surgery

Laryngeal carcinoma is the most common malignant tumor of the upper respiratory tract. Total laryngectomy provides complete and permanent detachment of the upper and lower airways that causes the loss of voice, leading to a patient's inability to verbally communicate in the postoperative period. This paper aims to exploit modern areas of deep learning research to objectively classify, extract and measure the substitution voicing after laryngeal oncosurgery from the audio signal. We propose using well-known convolutional neural networks (CNNs) applied for image classification for the analysis of voice audio signal. Our approach takes an input of Mel-frequency spectrogram (MFCC) as an input of deep neural network architecture. A database of digital speech recordings of 367 male subjects (279 normal speech samples and 88 pathological speech samples) was used. Our approach has shown the best true-positive rate of any of the compared state-of-the-art approaches, achieving an overall accuracy of 89.47%.