Validity, reliability and reproducibility of the VLS parameters form for the collection of videolaryngostroboscopic basic findings.

OBJECTIVE: The videolaryngostroboscopy parameters form (VLSP form) is a diagnostic tool for the collection of videolaryngostroboscopic basic findings through the evaluation of 12 parameters. The aim of the present study is to preliminarily investigate intra- and inter-rater reliability, validity and responsiveness of the VLSP form. METHODS: A study on a total amount of 160 forms for the evaluation of VLS basic findings was carried out. 80 forms were scored through the VLSP form and 80 with the Voice Vibratory Assessment with Laryngeal Imaging (VALI) form Stroboscopy (S) by four expert phoniatricians, that blindly scored the VLS recordings of 5 subjects without voice disorders and 5 patients with organic voice disorder before and after successful phonosurgery. Intra-rater and inter-rater analysis have been performed for both forms. The scores obtained through VLSP form and VALI form S have been compared to analyse concurrent validity, while VLSP scores before and after phonosurgery have been compared to analyse responsiveness. Finally, each rater annotated the "difficulty" in rating every parameter and its "importance" for the diagnosis. RESULTS: The VLSP form showed good inter- and intra-rater reliability. It showed a good accuracy for the documentation of changes of laryngeal anatomy and function after phonosurgery, similarly to the VALI form S. The 12 parameters of the VLSP form were judged "Slightly Important" in 28.3% of the samples, "Very Important" in 64.8% of the samples, "Not Difficult" in 73.1% of the samples. CONCLUSIONS: The results of the present study suggest that the VLSP form is comparable to the VALI form S for the evaluation of videolaryngostroboscopic parameters and is a valid, reliable and reproducible diagnostic tool. It can help voice clinicians in the evaluation of VLS examinations and it allows for a punctual assessment of modifications in laryngeal anatomy and function in pathological conditions and after phonosurgery.

Screening of voice disorders concerning anatomy teachers who use formaldehyde cadaveric parts in laboratory classes.

Voice disorders are common among teachers and, in particular, anatomy teachers are exposed to a potential enemy for dysphonia, irritating chemicals, that is, formaldehyde. We seek to verify the association between: (1) teaching time, (2) type of cadaveric conservation to which the teacher is exposed and (3) hours of exposure to cadaveric preservative related to the different categories of voice disorders screening (ITDV). The sample consisted of 111 teachers who answered to 02 data collection instruments: I - Sociodemographic Data; II - ITDV. Among participating teachers there were 71 male and 40 female, with an average age of 43 years and 11 months and an average teaching time of 16 years and 5 months. Association tests between teaching time and ITDV demonstrate a significant result in the relationship between voice failure and teaching time (p<0.05). All 111 teachers use their voices in laboratory classes and use cadaveric material. From those, 107 teachers are exposed to formaldehyde as cadaveric parts' conservative solution. There was a significant association (p<0.05) between voice failure and the type of cadaveric conservative solution but non-significant relationship (p>0.05) between ITDV and the time of exposure to formaldehyde preservative. Teachers' ITDV showed vocal signs and symptoms. In particular, voice loss due to time of teaching in anatomy, and voice failure, due to exposure to formaldehyde and combinations used in anatomical parts and cadavers, were significant.

Prevalence and Predictors of Post-Thyroidectomy Voice Dysfunction in a Nigerian Cohort.

BACKGROUND: Advances in surgical techniques have significantly improved the safety of thyroidectomy. Quality of life issues, such as voice changes, however continue to be important considerations. This study evaluated the prevalence and determinants of voice changes following thyroidectomies for non-malignant goiters in a Nigerian Hospital. METHODOLOGY: Consecutive adults who had thyroidectomy for non-malignant goiters were evaluated. Preoperatively, each participant had laryngoscopy and voice assessment using the Voice Handicap Index-10. At surgery, the intubation difficulty score, size of endotracheal tubes, and duration of operation were recorded. Laryngoscopy and VHI-10 assessments were repeated 1 week, 1 month, and 3 months after surgery. Pre and postoperative assessments were compared. RESULT: Fifty-four patients completed the study. Subtotal thyroidectomy was the commonest operation performed in 17 patients (31.48%), with 14 (25.93%) having total thyroidectomy. Baseline median VHI-10 score was 6.5(IQR-10). Five patients had abnormal preoperative laryngoscopy findings (benign vocal cord lesions). Postoperatively, seven patients had endoscopic evidence of vocal cord paralysis, six were unilateral while one was bilateral. Ten patients (18.5%) had early voice changes. Median VHI-10 score at one week was significantly higher than baseline (10 vs. 6.5, p = 0.01). At 3 months, VHI-10 reverted back to baseline in five of the ten patients while five patients (9.3%) had persistent voice changes. Abnormal endoscopic findings predicted the occurrence of early postoperative voice changes. CONCLUSION: Thyroidectomy for non-malignant goiters is associated with transient voice changes, many of which revert to normal over time. Abnormal endoscopic findings predict its occurrence.

Update on the pediatric adverse vocal behavior voice disorders: a clinical practice review.

The main causes of voice disorders in children with adverse vocal behavior include benign lesions of the vocal folds caused by voice abuse or misuses, such as vocal fold nodules, vocal fold polyps, and laryngitis. Long-term voice disorders can affect the physical and mental health of children. Reviewing the literature of the last two decades on "Vocal Fold Nodules," "Vocal Fold Polyp," "Voice disorder," "Voice Abuse," "Voice Misuse," "Pediatrics," and "Children" with the appropriate Boolean operators.  Conclusion: A total of 315 results were returned on an initial PubMed search. All articles from 2000 to 2022 written in English or Chinese were screened. Duplicate articles, those relating to adults only or concerned with the malignant lesion of the vocal cord, were excluded, resulting in 196 articles of interest. Relevant references and books have also been consulted, and we provide a review of the pathogenesis, diagnosis, and treatment of these maladaptive vocal behavioral voice disorders. What is Known: • Hoarseness is the most common voice symptom in children, and there are various causes of hoarseness in children. However, there is a lack of reviews on voice disorders caused by adverse vocal habits in children. • Voice training is a conservative treatment method for children with voice disorders , and it is important to clarify the factors that influence the effectiveness of voice training for children. What is New: • This review of the personality and family characteristics of children with adverse vocal behavioural voice disorders provides a valuable guide to the clinical planning of subsequent treatment. • This article discusses and summarises some of the factors that may influence the effectiveness of voice training in children and collates some of the scales and questionnaires currently used in children that are important in predicting the effectiveness of voice training.

Persistent subjective voice symptoms for two years after thyroidectomy.

BACKGROUND: Voice change after thyroidectomy is an important issue in thyroid surgery. However, little is known about long-term voice outcomes after thyroidectomy. This study investigates the long-term voice outcomes of thyroidectomy up to two years after surgery. Also, we analyzed the pattern of recovery through acoustic tests over time. METHODS: We reviewed data from 168 patients who underwent thyroidectomy between January 2020 and August 2020 at a single institution. The Thyroidectomy-related Voice and Symptom Questionnaire (TVSQ) score and acoustic voice analysis results were examined preoperatively and postoperative one, three, and six months, and one and two years after surgery. We divided patients into two groups based on the TVSQ score (≥15 or <15) at two years postoperatively. We investigated the difference of acoustic characteristics between the two groups and analyzed correlations between acoustic parameters and various clinical and surgical factors. RESULTS: Voice parameters tended to recover, but some parameters and TVSQ scores exhibited deterioration two years after surgery. In the subgroups, among the many clinicopathologic factors examined, voice abuse history including professional voice users (p = 0.014), greater extent of thyroidectomy and neck dissection (p = 0.019, p = 0.029), and high pitch voice (F0; p = 0.005, SFF; p = 0.016) were associated with high TVSQ score at two years. CONCLUSIONS: Patients frequently experience voice discomfort after thyroidectomy. After surgery, voice abuse history including professional voice users, greater extent of surgery, and higher pitch voice are associated with worse voice quality and increased risk of persistent voice symptoms over the long-term.

The role of patient perceptions and hyperfunctional voice disorders in predicting voice therapy attendance.

OBJECTIVES/HYPOTHESIS: Attendance is essential to voice therapy's effectiveness in the treatment of voice and laryngeal disorders. With such high rates of drop-out and non-attendance, it is important to understand the factors that influence this behavior. This study sought to identify potential predictors of attendance to voice therapy at an interdisciplinary voice clinic. STUDY DESIGN: Single-institution retrospective cohort study. METHODS: In this retrospective cohort study, patients evaluated at an interdisciplinary voice clinic who received a referral for voice therapy were identified. Age, gender, voice-related diagnoses, Voice Handicap Index-10 scores, Reflux Symptom Index scores, and measures of patient perceptions (self-rated severity, importance of voice in one's life, and "feelings about voice therapy") were recorded to evaluate associations with attendance to at least one therapy session. Standard statistical analysis and logistic regressions were performed. RESULTS: Of 168 subjects included, 111 (66.1 %) attended at least one session of voice therapy. Patients diagnosed with primary hyperfunctional voice disorders had a significantly higher attendance rate than other groups. Attenders had higher self-ratings of severity and more positive "feelings about voice therapy" compared to non-attenders. Regression models found three significant predictors of therapy attendance: primary diagnosis of hyperfunctional voice disorder, self-rated severity, and "feelings about voice therapy." CONCLUSION: In this cohort, patients with more positive feelings about voice therapy, higher self-rated severity, and a diagnosis of primary hyperfunctional voice disorder were more likely to attend voice therapy.

Presbylarynx: validation of a classification based on morphological characteristics.

PURPOSE: Validation of a classification to stratify presbylarynx into three types. METHODS: Evaluate the reproducibility of this classification. 30 video recordings of rigid laryngostroboscopies were presented to 20 otorhinolaryngologists from five tertiary hospitals: 10 residents and 10 specialists. Specialists with different differentiations were included. RESULTS: Among 1200 evaluations performed (30 video recordings × 20 raters, twice), average intra-rater agreement rate was 0.847 (p < 0.001). The average inter-rater reliability was 0.67 ± 0.179 on the first evaluation, and 0.691 ± 0.131 on the second evaluation. CONCLUSIONS: This study presented the validation of a classification that allows stratifying laryngeal endoscopic signs in elderly patients with presbyphonia. We believe that this classification will enrich the diagnostic protocol of the aging voice and will improve the documentation of treatment outcomes.

Pathological voice classification based on multi-domain features and deep hierarchical extreme learning machine.

The intelligent data-driven screening of pathological voice signals is a non-invasive and real-time tool for computer-aided diagnosis that has attracted increasing attention from researchers and clinicians. In this paper, the authors propose multi-domain features and the hierarchical extreme learning machine (H-ELM) for the automatic identification of voice disorders. A sufficient number of sensitive features are first extracted from the original voice signal through multi-domain feature extraction (i.e., features of the time domain and the sample entropy based on ensemble empirical mode decomposition and gammatone frequency cepstral coefficients). To eliminate redundancy in high-dimensional features, neighborhood component analysis is then applied to filter out sensitive features from the high-dimensional feature vectors to improve the efficiency of network training and reduce overfitting. The sensitive features thus obtained are then used to train the H-ELM for pathological voice classification. The results of the experiments showed that the sensitivity, specificity, F1 score, and accuracy of the H-ELM were 99.37%, 98.61%, 99.37%, and 98.99%, respectively. Therefore, the proposed method is feasible for the initial classification of pathological voice signals.

Amateur choir singers - Does good vocal health matter?

BACKGROUND: Many amateur singers enjoy singing in choirs. They are likely to lack the training and expertise of professional singers and this may have an impact on their vocal health. AIMS: To assess the experiences of amateur singers, their use of warm-up and cool-down sessions, their vocal health, their sources of help and advice and their reactions to receiving a diagnosis and treatment. METHODS AND PROCEDURES: A questionnaire was sent to amateur choir singers. Quantitative data on their use of warm-up and cool-down activities were collected and the singers rated how frequently they experienced adverse voice symptoms after singing. Qualitative data were collected from singers who had sought advice about their voice and from those who had received a diagnosis and treatment. OUTCOME AND RESULTS: Most choirs used warm-up sessions but few used cool-down. Singers who participated in choir warm-ups experienced significantly less vocal symptoms. Individual warm-ups were ineffective. Singers who had previously had a diagnosis continued to have significantly higher symptom scores. A quarter of the singers who had neither sought help nor had a diagnosis had high scores. Singers appeared uncertain about who to consult for help. Those who saw a speech and language therapist were much more likely to receive a diagnosis and treatment either by the therapist or by referral to an ear, nose and throat specialist. Those receiving treatment were mainly positive about it and likely to complete a course of therapy. CONCLUSIONS AND IMPLICATIONS: Warm-up sessions performed by choirs are helpful in avoiding voice symptoms and are recommended. The presence of singers who are unaware of potential damage to their voices is a concern. A simple rating system, as used in this research, may act to alert them to the dangers. Choir leaders could be more active in advising singers and refer them to professionals where necessary. Speech and language therapists appear more sensitive to their problems and could be a first point of contact. WHAT THIS PAPER ADDS: What is already known on this subject Poor vocal health may occur in amateur singers and lead to voice disorders which will reduce the demonstrated benefits of regular singing. They may be less aware than professional singers of potential hazards and may not know how to sing in a way that ensures good vocal health. There has been less research on the extent of voice problems in amateurs or of the actions they may take to resolve the vocal difficulties they encounter. What this paper adds to the existing knowledge This survey demonstrates that choir warm-ups were effective in helping to avoid common symptoms associated with voice problems. A large minority of singers who had not sought help for voice problems had high symptom scores on our assessment. Singers frequently turned to a singing teacher for help. Those who consulted a speech and language therapist were much more likely to be referred for further investigation and treatment. What are the potential or actual clinical implications of this work? We present a simple subjective measure of symptom severity which succeeds in distinguishing between singers who warm up their voices and those who do not and between singers who have received a past diagnosis and those who have not. There are a number of indications in this research that amateur singers are unsure if they need help and are uncertain who can offer it. In these circumstances the use of a simple measure of the severity of vocal abuse may be useful.

Prospective voice assessment after uncomplicated thyroidectomy: A comprehensive analysis of a single centre experience.

OBJECTIVES: Voice change after uncomplicated thyroidectomy has been an important issue in the field of thyroid surgery. The aim of this study was to promote understanding of voice change after uncomplicated thyroidectomy by analysing the results for a large number of patients from a single institute. DESIGN: We retrospectively reviewed the medical records of 2879 consecutive patients who underwent thyroidectomy and voice evaluation between January 2014 and December 2019 in a single institute. All the patients had their vocal status assessed using videostroboscopy, acoustic voice analyses, aerodynamic study, and Thyroidectomy-related Voice and Symptom Questionnaire (TVSQ) scores preoperatively and at 1, 3, and 6 months postoperatively. We analysed the pattern of voice changes over time and differences in voice parameters based on clinical factors. To confirm the usefulness of the TVSQ, the correlation between TVSQ scores and objective parameters was analysed. Lastly, predictive factors for persistent voice symptoms were analysed. SETTING: Tertiary referral hospital. RESULTS: The frequency ranges and TVSQ scores exhibited significant deterioration until 6 months following surgery. Among clinical factors, the extents of thyroidectomy and neck dissection were associated with worse voice parameters. The TVSQ score was significantly correlated with objective voice parameters. The extents of thyroidectomy and neck dissection were predictive of persistent voice symptoms at 6 months after thyroidectomy. CONCLUSION: After uncomplicated thyroidectomy, most voice parameters tended to recover, but some parameters remained aggravated even at 6 months after surgery. With more extensive surgery, worse voice quality and the higher risk of persistent voice symptoms may be anticipated.

[Patient-reported outcome measures for assessing health-related quality of life in patients with voice and swallowing disorders]. / Patient-Reported Outcome Measures zur Erfassung der gesundheitsbezogenen Lebensqualität bei Patienten mit Stimm- und Schluckstörungen.

Standardized and validated patient surveys in the form of questionnaires are becoming increasingly important in otorhinolaryngology. Subjective symptoms and functional limitations but also health-related quality of life are assessed with so-called patient-reported outcome measures (PROMs). Besides patient history and objective findings, these questionnaires can be applied as another important source for ensuring the quality of patient-centered care. In German-speaking countries there are now several PROMs that specifically ask for the patient perspective on voice and swallowing disorders. For patients with voice disorders the following questionnaires based on the Voice Handicap Index (VHI), which are also used internationally, are mainly applied: VHI-30, VHI-12i, and VHI-9i. In the field of oropharyngeal dysphagia, the German versions of the Eating Assessment Tool-10 (EAT-10), des Sydney Swallow Questionnaire SSQ­G, the Swallowing Quality of Life Questionnaire (SWAL-QoL), and the MD Anderson Dysphagia Inventory (MDADI) are currently mostly used.

Effect of Pompage on Vocal and Respiratory Measurements from Female Teachers: A Randomized Clinical Trial.

INTRODUCTION: Teachers stand out among occupational groups due to their contribution to modern societies, and their voice is the main form of interaction. OBJECTIVE: The aim of the study was to verify the changes in vocal and respiratory measurements from teachers with vocal and musculoskeletal complaints and with normal larynx after applying a musculoskeletal manipulation protocol of myofascial release using pompage. METHODS: Controlled and randomized clinical trial with 56 participants: 28 teachers in the study group and 28 teachers in the control group. Anamnesis, videolaryngoscopy, hearing screening, sound pressure and maximum phonation time measurements, and manovacuometry were performed. The musculoskeletal manipulation protocol of myofascial release using pompage consisted of a total of 24 sessions, 40 min for each session, three times a week, for 8 weeks. RESULTS: There was a significant improvement in the maximum respiratory pressure in the study group after the intervention. The sound pressure level and the maximum phonation time did not change significantly. DISCUSSION AND CONCLUSION: Musculoskeletal manipulation protocol of myofascial release using pompage had a direct effect on the respiratory measurements from female teachers, significantly increasing the maximum respiratory pressure but without affecting sound pressure level and the /a/ maximum phonation time.

Vocal Health Knowledge Status and Attitudes of Professional and Amateur Performing Artists in Greece.

OBJECTIVE: This study aimed to investigate the vocal health literacy level and attitudes of professional and amateur performing artists in Greece and identify possible gaps and limitations that should be addressed in the future. METHODS: This mixed-methods study took place in two successive stages. The first stage involved 4 focus groups with professional and amateur actors and singers. The focused discussions of this stage, which were moderated by one otolaryngologist and one speech and language therapist, provided necessary information for the design of the structured questionnaire used during the second stage of the study (online survey). The online questionnaire included questions about responders' knowledge and attitudes of vocal health. Although not compulsory, the Greek version of the Singing Voice Handicap Index (SVHI) was also part of the online survey. RESULTS: The online questionnaire was validly answered by 305 respondents during a 10-month period (May 2020 to March 2021). Considerable knowledge gaps were observed regarding anatomy, physiology, and vocal disorders management. The vast majority of participants (94.1%) had at some point experienced problems with their voice. Gastroesophageal reflux (21.6%), vocal nodules (12.5%), and vocal polyps (3.9%) were the most frequently mentioned diagnoses in our sample. A small proportion of participants (9.5%) stated that they have never visited an ENT, while 84.3% have never seen a speech and language therapist. The total scores on the SVHI in our sample were compatible to the ones in previous studies targeting healthy professional and amateur artists (range 0-116, mean 32, SD 24.7). CONCLUSIONS: This study provides insights into Greek performing artists' knowledge and attitudes regarding vocal health and may pave the way towards improving vocal health awareness among performing artists.

VOICE CONDITION FOLLOWING SHORT-TERM ENDOTRACHEAL INTUBATION IN HEAD AND NECK SURGERY: STUDY PROTOCOL FOR CLINICAL TRIAL.

Endotracheal intubation is an airway management procedure commonly performed under general anesthesia. It is linked with postoperative voice changes. The incidence and reasons of hoarseness and vocal cord injury are not very well investigated, especially after short-term anesthesia and intubation in head and neck surgery. The aim of the study is to identify the causes of voice changes after short endotracheal anesthesia in head and neck surgery. The study will include patients scheduled for head and neck surgery under general anesthesia with endotracheal intubation up to 3 hours. There will be 3 groups of patients, as follows: thyroid surgery, non-thyroid surgery, and control group undergoing surgery outside head and neck. Videostroboscopy will be recorded before and after surgery. Further diagnostic workup will include voice status; subjective voice self-analysis; perceptive and objective acoustic voice analysis at 4 time points (preoperatively, postoperative day 2, 2 weeks and 1 month after surgery). Endotracheal intubation is a safe method of airway management although it can temporarily alter a patient's voice quality. It is not known how much of this is the result of anesthesia, general condition of the patient, or surgery. This trial is expected to shed some light on this issue.

Parkinson's disease patients with freezing of gait have more severe voice impairment than non-freezers during "ON state".

BACKGROUND: Speech disorders and freezing of gait (FOG) in Parkinson's disease (PD) may have some common pathological mechanisms. The purpose of this study was to compare the acoustic parameters of PD patients with dopamine-responsive FOG (PD-FOG) and without FOG (PD-nFOG) during "ON state" and explore the ability of "ON state" voice features in distinguishing PD-FOG from PD-nFOG. METHODS: A total of 120 subjects, including 40 PD patients with dopamine-responsive FOG, 40 PD-nFOG, and 40 healthy controls (HCs) were recruited. All subjects underwent neuropsychological tests. Speech samples were recorded through the sustained vowel pronunciation tasks during the "ON state" and then analyzed by the Praat software. A set of 27 voice features was extracted from each sample for comparison. Support vector machine (SVM) was used to build mathematical models to classify PD-FOG and PD-nFOG. RESULTS: Compared with PD-nFOG, the jitter, the standard deviation of fundamental frequency (F0SD), the standard deviation of pulse period (pulse period SD) and the noise-homophonic-ratio (NHR) were increased, and the maximum phonation time (MPT) was decreased in PD-FOG. The above voice features were correlated with the freezing of gait questionnaire (FOGQ). The average accuracy, specificity, and sensitivity of SVM models based on 27 voice features for classifying PD-FOG and PD-nFOG were 73.57%, 75.71%, and 71.43%, respectively. CONCLUSIONS: PD-FOG have more severe voice impairment than PD-nFOG during "ON state".

The Effectiveness of Supervised Machine Learning in Screening and Diagnosing Voice Disorders: Systematic Review and Meta-analysis.

BACKGROUND: When investigating voice disorders a series of processes are used when including voice screening and diagnosis. Both methods have limited standardized tests, which are affected by the clinician's experience and subjective judgment. Machine learning (ML) algorithms have been used as an objective tool in screening or diagnosing voice disorders. However, the effectiveness of ML algorithms in assessing and diagnosing voice disorders has not received sufficient scholarly attention. OBJECTIVE: This systematic review aimed to assess the effectiveness of ML algorithms in screening and diagnosing voice disorders. METHODS: An electronic search was conducted in 5 databases. Studies that examined the performance (accuracy, sensitivity, and specificity) of any ML algorithm in detecting pathological voice samples were included. Two reviewers independently selected the studies, extracted data from the included studies, and assessed the risk of bias. The methodological quality of each study was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 tool via RevMan 5 software (Cochrane Library). The characteristics of studies, population, and index tests were extracted, and meta-analyses were conducted to pool the accuracy, sensitivity, and specificity of ML techniques. The issue of heterogeneity was addressed by discussing possible sources and excluding studies when necessary. RESULTS: Of the 1409 records retrieved, 13 studies and 4079 participants were included in this review. A total of 13 ML techniques were used in the included studies, with the most common technique being least squares support vector machine. The pooled accuracy, sensitivity, and specificity of ML techniques in screening voice disorders were 93%, 96%, and 93%, respectively. Least squares support vector machine had the highest accuracy (99%), while the K-nearest neighbor algorithm had the highest sensitivity (98%) and specificity (98%). Quadric discriminant analysis achieved the lowest accuracy (91%), sensitivity (89%), and specificity (89%). CONCLUSIONS: ML showed promising findings in the screening of voice disorders. However, the findings were not conclusive in diagnosing voice disorders owing to the limited number of studies that used ML for diagnostic purposes; thus, more investigations are needed. While it might not be possible to use ML alone as a substitute for current diagnostic tools, it may be used as a decision support tool for clinicians to assess their patients, which could improve the management process for assessment. TRIAL REGISTRATION: PROSPERO CRD42020214438; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=214438.

Development of a machine-learning based voice disorder screening tool.

OBJECTIVE: Early recognition and referral are crucial for voice disorder management. Limited availability of subspecialists, poor primary care awareness, and the need for specialized equipment impede effective care. Thus, there is a need for a tool to improve voice pathology screening. Machine learning algorithms (MLAs) have shown promise in analyzing acoustic characteristics of phonation. However, few studies report clinical applications of MLAs for voice pathology detection. The objective of this study was to design and validate a MLA for detecting pathological voices. METHODS: A MLA was developed for voice analysis. Audio samples converted into spectrograms were inputted into a pre-existing VGG19 convolutional neural network (CNN) and image-classifier. The resulting feature map was classified as either pathological or healthy using a Support Vector Machine (SVM) binary linear classifier. This combined MLA was "trained" with 950 sustained "/i/" vowel audio samples from the Saarbrucken Voice Database (SVD), which contains subjects with and without voice disorders. The trained MLA was "tested" with 406 SVD samples to determine sensitivity, specificity, and overall accuracy. External validation of the MLA was performed using clinical voice samples collected from patients attending a subspecialty voice clinic. RESULTS: The MLA detected pathologies in SVD samples with 98.5% sensitivity, 97.1% specificity and 97.8% overall accuracy. In 30 samples obtained prospectively from voice clinic patients, the MLA detected pathologies with 100% sensitivity, 96.3% specificity and 96.7% overall accuracy. CONCLUSIONS: This study demonstrates that a MLA using a simple audio input can detect diverse vocal pathologies with high sensitivity and specificity. Thus, this algorithm shows promise as a potential screening tool.

Multiparametric analysis of voice following prolonged voice use and voice rest in teachers: evidence from discriminant analysis.

PURPOSE: Even though recent investigations have used multiparametric protocol, the set of robust parameters in determining the effects of vocal fatigue and voice rest in teachers is not clear. The first objective of the study was to document the impact of prolonged voice use and voice rest on the subjective and objective voice parameters among Indian secondary school teachers. The second objective was to determine the set voice parameters sensitive to vocal changes resulting from continuous voice use and voice rest. METHOD: The study included 15 male and 15 female secondary school teachers with a clinically normal voice and no history of voice disorders. Phonation samples were recorded in three different conditions, i.e., condition 1 (before voice use), condition 2 (following voice use), and condition 3 (following voice rest). The vocal Fatigue Index (version 2) was administered before the voice recordings in all three conditions. The objective parameters, namely fundamental frequency, range of fundamental frequency, jitter (%), shimmer (%), harmonic to noise ratio, and smoothened cepstral peak prominence, were extracted. RESULTS: Results revealed that fundamental frequency, jitter, shimmer, Harmonic to noise ratio, and smoothened cepstral peak prominence were significantly different across the three conditions. The discriminant analysis revealed that only three parameters classified 98.3% of samples accurately between the three conditions. CONCLUSION: Further research on the correlation between the other subjective and the objective parameters of voice after vocal fatigue would provide more penetrating and ample in-depth insights into the assessment and quantification of vocal fatigue.

Schoolteachers with voice handicap are twice as likely to report depressive symptoms.

This study aimed to examine the association between voice disorder and depressive symptoms in schoolteachers. A cross-sectional survey was conducted with elementary and secondary schoolteachers. Voice disorders and depressive symptoms were assessed with the Voice Handicap Index-10 and the Beck Depression Inventory-II, respectively. Logistic and linear regressions models were adjusted for confounding variables. In the 389 schoolteachers studied, the prevalence of voice handicap and depressive symptoms was 18.8 and 38.8%, respectively. Voice handicap was associated with depressive symptoms on all models tested. The logistic regression showed an odds ratio of 2.21 (95% confidence interval: 1.19, 4.08; p value < 0.05), while in the linear regression each point increase on the voice disorder scale increased the Beck Depression Inventory-II score by 0.39 points (95% confidence interval: 0.26, 0.54; p value < 0.05). This study showed that teachers of public schools with voice handicap are twice as likely to report depressive symptoms.

Using SincNet for Learning Pathological Voice Disorders.

Deep learning techniques such as convolutional neural networks (CNN) have been successfully applied to identify pathological voices. However, the major disadvantage of using these advanced models is the lack of interpretability in explaining the predicted outcomes. This drawback further introduces a bottleneck for promoting the classification or detection of voice-disorder systems, especially in this pandemic period. In this paper, we proposed using a series of learnable sinc functions to replace the very first layer of a commonly used CNN to develop an explainable SincNet system for classifying or detecting pathological voices. The applied sinc filters, a front-end signal processor in SincNet, are critical for constructing the meaningful layer and are directly used to extract the acoustic features for following networks to generate high-level voice information. We conducted our tests on three different Far Eastern Memorial Hospital voice datasets. From our evaluations, the proposed approach achieves the highest 7%-accuracy and 9%-sensitivity improvements from conventional methods and thus demonstrates superior performance in predicting input pathological waveforms of the SincNet system. More importantly, we intended to give possible explanations between the system output and the first-layer extracted speech features based on our evaluated results.