Temporal and Spectral Models as Correlates to Auditory-Perceptual Judgments of Overall Severity and Listener Comfort in Tracheoesophageal Voice.

INTRODUCTION: This study pursued two objectives: (1) to determine the potential association between listener (n = 51) judgments of 20 male tracheoesophageal speaker samples for two auditory-perceptual dimensions of voice, overall severity (OS) and listener comfort (LC); and (2) to assess the temporal and spectral acoustic correlates for these auditory-perceptual dimensions. METHODOLOGY: Three separate correlation analyses were performed to evaluate the association between OS and LC. First, scores of OS and LC from all listeners were pooled together, and then the correlation between OS and LC was computed. Second, scores of OS and LC were averaged over all listeners to derive a single estimate of OS and LC for each TE speaker sample; the correlation between the average OS and LC was then computed. Third, listener-to-listener variability in the association between OS and LC was evaluated by computing the correlation between OS and LC scores from each listener across all TE samples. Finally, two stepwise multiple regression models were created to relate the average LC score to spectral and temporal variation in the acoustic signal. RESULTS: While the pooled OS and LC scores had a moderate positive correlation (r = 0.66, p < 0.00001), the averaged OS and LC exhibited a near perfect positive correlation (r = 0.99, p < 0.00001). The significant differences between the pooled and averaged scores were explained by significant listener-to-listener variability in the association between OS and LC. OS and LC scores from 5 listeners had non-significant correlations, 10 had moderate correlations (r < 0.7), 35 listeners had high correlations (0.7 < r < 0.9), and 1 listener had a very high correlation (r < 0.9 < 1). Finally, the acoustic models created based on the spectral and temporal variations in the signal were able to account for 87.7% and 61.8% of variation in the average LC score. CONCLUSIONS: The strong correlations between OS and LC suggest that LC may, in fact, provide a more comprehensive auditory-perceptual surrogate for the voice quality of TE speakers. Although OS and LC are distinct conceptual dimensions, LC appears to have the advantage of assessing the social impact and potential communication disability that may exist in interactions between TE speakers and listeners.

A Scoping Review of Artificial Intelligence Detection of Voice Pathology: Challenges and Opportunities.

OBJECTIVE: Survey the current literature on artificial intelligence (AI) applications for detecting and classifying vocal pathology using voice recordings, and identify challenges and opportunities for advancing the field forward. DATA SOURCES: PubMed, EMBASE, CINAHL, and Scopus databases. REVIEW METHODS: A comprehensive literature search was performed following the Preferred Reporting Items for Systematic Reviews and Meta-analyses Extension for Scoping Reviews guidelines. Peer-reviewed journal articles in the English language were included if they used an AI approach to detect or classify pathological voices using voice recordings from patients diagnosed with vocal pathologies. RESULTS: Eighty-two studies were included in the review between the years 2000 and 2023, with an increase in publication rate from one study per year in 2012 to 10 per year in 2022. Seventy-two studies (88%) were aimed at detecting the presence of voice pathology, 24 (29%) at classifying the type of voice pathology present, and 4 (5%) at assessing pathological voice using the Grade, Roughness, Breathiness, Asthenia, and Strain scale. Thirty-six databases were used to collect and analyze speech samples. Fourteen articles (17%) did not provide information about their AI model validation methodology. Zero studies moved beyond the preclinical and offline AI model development stages. Zero studies specified following a reporting guideline for AI research. CONCLUSION: There is rising interest in the potential of AI technology to aid the detection and classification of voice pathology. Three challenges-and areas of opportunities-for advancing this research are heterogeneity of databases, lack of clinical validation studies, and inconsistent reporting.

Sarcopenia as a Predictor of Feeding Tube Placement in Individuals with Oropharyngeal Cancer.

Purpose: In oropharyngeal squamous cell carcinoma (OPSCC), systemic loss of skeletal muscle mass (SMM), or sarcopenia, is a strong prognostic predictor of survival outcomes. However, the relationship between sarcopenia and nutrition-related outcomes is not well understood. This investigation evaluated the prognostic significance of sarcopenia for feeding tube (FT) placement in a cohort of OPSCC patients. Methods and Materials: A retrospective cohort study was conducted with data collected from 194 OPSCC patients treated with definitive radiation therapy (RT) or chemoradiation therapy (CRT). Sarcopenia was assessed from computed tomography imaging at the level of the third cervical (C3) and fourth thoracic (T4) vertebrae. The prognostic nature of pretreatment sarcopenia and its relationship with FT placement was explored using logistic regression. Results: The median age of patients included was 61.0 years, and the majority were male (83%). In this patient cohort, 87.6% underwent concurrent CRT, and 30.9% received a FT over the course of treatment. Sarcopenia was identified at baseline in 72.7% of patients based on C3 SMM measurements and in 41.7% based on measures at the level of T4. Based on measures at both C3 and T4, those with sarcopenia were significantly more likely to receive a FT and had significantly worse freedom from FT placement compared with patients without sarcopenia. Sarcopenia assessed at T4 was a significant predictor of FT placement. Conclusions: SMM measured at T4 may represent a novel and practical biomarker for sarcopenia detection that is associated with the need for FT placement. These findings suggest that the detection of baseline sarcopenia could guide decision-making related to the need for nutritional support in OPSCC patients undergoing RT/CRT.

The Effect of Rating Method on Reliability of Judgments of Strain Across Populations.

PURPOSE: Variability in auditory-perceptual ratings of voice limits their utility, with the poorest reliability often noted for vocal strain. The purpose of this study was to determine whether an experimental method, called visual sort and rate (VSR), promoted stronger rater reliability than visual analog scale (VAS), for ratings of strain in two clinical populations: adductor laryngeal dystonia (ADLD) and vocal hyperfunction (VH). METHOD: Connected speech samples from speakers with ADLD and VH as well as age- and sex-matched controls were selected from a database. Fifteen inexperienced listeners rated strain for two speaker sets (25 ADLD speakers and five controls; 25 VH speakers and five controls) across four rating blocks: VAS-ADLD, VSR-ADLD, VAS-VH, and VSR-VH. For the VAS task, listeners rated each speaker for strain using a vertically oriented 100-mm VAS. For the VSR task, stimuli were distributed into sets of samples with a range of severities in each set. Listeners sorted and ranked samples for strain within each set, and final ratings were captured on a vertically oriented 100-mm VAS. Intrarater reliability (Pearson's r) and interrater variability (mean of the squared differences between a listener's ratings and group mean ratings) were compared across rating methods and populations using two repeated-measures analyses of variance. RESULTS: Intrarater reliability of strain was significantly stronger when listeners used VSR compared to VAS; listeners also showed significantly better intrarater reliability in ADLD than VH. Listeners demonstrated significantly less interrater variability (better reliability) when using VSR compared to VAS. No significant effect of population or interactions was found between listeners for measures of interrater variability. CONCLUSIONS: VSR increases intrarater reliability for ratings of vocal strain in speakers with VH and ADLD. VSR decreases variability of auditory-perceptual judgments of strain between inexperienced listeners in these clinical populations. Future research should determine whether benefits of VSR extend to voice clinicians and/or clinical settings.

Perceptions and practices of people with a total laryngectomy during COVID-19 pandemic: A mixed methods analysis.

PURPOSE: People with a total laryngectomy (PTL) confront safety threats related to altered airway anatomy and risk of adverse events is amplified during healthcare crises, as exemplified by COVID-19 pandemic. Understanding these challenges, how they are navigated by PTL, and what resources can be deployed to alleviate risk can improve interprofessional care by speech-language pathologists (SLPs), otolaryngologists, and other professionals. MATERIALS AND METHODS: An online survey was disseminated to PTL in the United States during the COVID-19 pandemic, querying participants about safety concerns and sources of information accessed to address care. Descriptive statistics and Chi-square were used to analyze information sources consumed by tracheoesophageal, esophageal, and electrolaryngeal speakers. Content analysis was completed to identify themes and quantify responses by subtheme. RESULTS: Among 173 respondent PTL, tracheoesophageal speakers preferentially sought otolaryngologist input, whereas esophageal and electrolaryngeal speakers more often chose SLPs (p < .01). Overall, tracheoesophageal speakers had more SLP or otolaryngologist contact. Many PTL reported stringent handwashing, neck cleaning, and hygienic risk mitigation strategies. Six themes emerged in content analysis involving risk of infection/transmission, heightened vigilance, changes to alaryngeal communication, modified tracheostoma coverage, diagnostic testing, and risk from comorbid conditions. Limited provider contact suggested pandemic barriers to healthcare access. CONCLUSIONS: PTL have a range of laryngectomy-specific needs and concerns, and type of alaryngeal communication was associated with source of information sought. Collaborations among healthcare professionals need to be optimized to improve patient navigation and overall access to specialized care.

End-to-end deep learning classification of vocal pathology using stacked vowels.

Objectives: Advances in artificial intelligence (AI) technology have increased the feasibility of classifying voice disorders using voice recordings as a screening tool. This work develops upon previous models that take in single vowel recordings by analyzing multiple vowel recordings simultaneously to enhance prediction of vocal pathology. Methods: Voice samples from the Saarbruecken Voice Database, including three sustained vowels (/a/, /i/, /u/) from 687 healthy human participants and 334 dysphonic patients, were used to train 1-dimensional convolutional neural network models for multiclass classification of healthy, hyperfunctional dysphonia, and laryngitis voice recordings. Three models were trained: (1) a baseline model that analyzed individual vowels in isolation, (2) a stacked vowel model that analyzed three vowels (/a/, /i/, /u/) in the neutral pitch simultaneously, and (3) a stacked pitch model that analyzed the /a/ vowel in three pitches (low, neutral, and high) simultaneously. Results: For multiclass classification of healthy, hyperfunctional dysphonia, and laryngitis voice recordings, the stacked vowel model demonstrated higher performance compared with the baseline and stacked pitch models (F1 score 0.81 vs. 0.77 and 0.78, respectively). Specifically, the stacked vowel model achieved higher performance for class-specific classification of hyperfunctional dysphonia voice samples compared with the baseline and stacked pitch models (F1 score 0.56 vs. 0.49 and 0.50, respectively). Conclusions: This study demonstrates the feasibility and potential of analyzing multiple sustained vowel recordings simultaneously to improve AI-driven screening and classification of vocal pathology. The stacked vowel model architecture in particular offers promise to enhance such an approach. Lay Summary: AI analysis of multiple vowel recordings can improve classification of voice pathologies compared with models using a single sustained vowel and offer a strategy to enhance AI-driven screening of voice disorders. Level of Evidence: 3.

Self-perception of fatigue in individuals diagnosed with head and neck cancer.

PURPOSE: Head and neck cancer (HNCa) presents numerous challenges secondary to treatment. While there is substantial clinical awareness to the range of challenges demonstrated in this population, information on the impact of post-treatment fatigue is limited. This study investigated the degree of perceived fatigue in those treated for HNCa. METHODS: The study was a cross-sectional, self-report, survey design. Adult participants (n = 47) completed a series of three questionnaires; two validated fatigue measures - the Fatigue Screening Inventory (FSI) and the Multidimensional Fatigue Inventory (MFI-20) and a general health-related quality of life measure the European Organisation of Research on the Treatment of Cancer - Quality of Life Questionnaire (EORTC-QLQC30) and the head and neck site specific module (QLQ - H&N 35) were administered. RESULTS: Of the 47 participants, more than half (55%) were identified as having clinically significant self-reported levels of fatigue. Correlational analysis revealed an inverse relationship between fatigue and overall health-related quality of life (HRQOL) implying that as fatigue increases, one's perceived HRQOL decreases. CONCLUSIONS: These data suggest that efforts to proactively screen for and index fatigue and seek anticipatory interventions may benefit both short- and long-term HRQOL outcomes in those diagnosed with HNCa. LEVEL OF EVIDENCE: IV.

Auditory-Perceptual and Pupillometric Evaluation of Vocal Roughness and Listening Effort in Tracheoesophageal Speech.

OBJECTIVES: This study evaluated auditory-perceptual judgments of perceived vocal roughness (VR) and listening effort (LE) along with pupillometric responses in response to speech samples produced by tracheoesophageal (TE) talkers. METHODS: Twenty normal-hearing, naive young adults (eight men and twelve women) served as listeners. Listeners were divided into two groups: (1) a with-anchor (WA) group (four men and six women) and (2) a no-anchor (NA) group (four men and six women). All were presented with speech samples produced by twenty TE talkers; listeners evaluated two auditory-perceptual dimensions-VR and LE-using visual analog scales. Anchors were provided to the WA group as an external referent for their ratings. In addition, during the auditory-perceptual task, each listener's pupil reactions also were recorded with peak pupil dilation (PPD) measures extracted as a physiologic indicator associated with the listening task. RESULTS: High interrater reliability was obtained for both the WA and NA groups. High correlations also were observed between auditory-perceptual ratings of roughness and LE, and between PPD values and ratings of both dimensions for the WA group. The inclusion of an anchor during the auditory-perceptual task improved interrater reliability ratings, but it also imposed an increased demand on listeners. CONCLUSIONS: Data obtained offer insights into the relationship between subjective indices of voice quality (ie, auditory-perceptual evaluation) and physiologic responses (PPD) to the abnormal voice quality that characterizes TE talkers. Furthermore, these data provide information on the inclusion/exclusion of audio anchors and potential increases in listener demand in response to abnormal voice quality.

Adoption and Utilization of Heat and Moisture Exchangers (HMEs) in the Tracheostomy Patient.

OBJECTIVE: Management of tracheostomized patients typically involves a conventional external humidification system (CEHS). CEHS are noisy, negatively impact patient mobility, and increases costs. Additionally, they prevent phonation and the ability to cough. Alternatively, heat and moisture exchange (HME) devices have been used in laryngectomized patients. We present an institutional quality improvement project exploring the use and efficacy of an HME device following tracheostomy. METHODS: Health care professionals and stakeholders from multiple disciplines were identified: otolaryngology, nursing, administration, case management, and speech-language pathology. The focus was on an otolaryngology acute care nursing unit. Protocols for product acquisition, nursing education, care flowcharts, and discharge planning were established. Efficacy was assessed by tracking patient pulmonary status, nursing notes, and questionnaires. RESULTS: Seventy-one tracheostomized patients were enrolled. Two patients (2.8%) were unable to tolerate the HME. There were no complications from mucous plugging or respiratory distress. Eighty-nine percent of nursing staff surveyed preferred the use of an HME device over CEHS, particularly for ease of patient mobility. Additional favorable findings were patient satisfaction, cost savings, reduced noise, communication, and ease of discharge education and planning. DISCUSSION: Replacing CEHS with HMEs provides distinct advantages, with a positive impact on patients, family members, and health care personnel. Resistance to changing from the traditional standard of care was alleviated with education, focused training, and positive outcomes. IMPLICATIONS FOR PRACTICE: These data indicate that an HME device is safe and offers advantages to both patients and nurses over traditional CEHS.

Seroprevalence of anti-Lassa Virus IgG antibodies in three districts of Sierra Leone: A cross-sectional, population-based study.

BACKGROUND: Lassa virus (LASV), the cause of the acute viral hemorrhagic illness Lassa fever (LF), is endemic in West Africa. Infections in humans occur mainly after exposure to infected excrement or urine of the rodent-host, Mastomys natalensis. The prevalence of exposure to LASV in Sierra Leone is crudely estimated and largely unknown. This cross-sectional study aimed to establish a baseline point seroprevalence of IgG antibodies to LASV in three administrative districts of Sierra Leone and identify potential risk factors for seropositivity and LASV exposure. METHODOLOGY AND PRINCIPAL FINDINGS: Between 2015 and 2018, over 10,642 participants from Kenema, Tonkolili, and Port Loko Districts were enrolled in this cross-sectional study. Previous LASV and LF epidemiological studies support classification of these districts as "endemic," "emerging," and "non-endemic", respectively. Dried blood spot samples were tested for LASV antibodies by ELISA to determine the seropositivity of participants, indicating previous exposure to LASV. Surveys were administered to each participant to assess demographic and environmental factors associated with a higher risk of exposure to LASV. Overall seroprevalence for antibodies to LASV was 16.0%. In Kenema, Port Loko, and Tonkolili Districts, seroprevalences were 20.1%, 14.1%, and 10.6%, respectively. In a multivariate analysis, individuals were more likely to be LASV seropositive if they were living in Kenema District, regardless of sex, age, or occupation. Environmental factors contributed to an increased risk of LASV exposure, including poor housing construction and proximity to bushland, forested areas, and refuse. CONCLUSIONS AND SIGNIFICANCE: In this study we determine a baseline LASV seroprevalence in three districts which will inform future epidemiological, ecological, and clinical studies on LF and the LASV in Sierra Leone. The heterogeneity of the distribution of LASV and LF over both space, and time, can make the design of efficacy trials and intervention programs difficult. Having more studies on the prevalence of LASV and identifying potential hyper-endemic areas will greatly increase the awareness of LF and improve targeted control programs related to LASV.