Journal of Voice最新文献

{"title":"Human Vocal Fold Tissue Modifications Related to Laryngopharyngeal Reflux Disease: A Systematic Review.","authors":"Guangjin Chen, Jérôme R Lechien","doi":"10.1016/j.jvoice.2025.04.019","DOIUrl":"https://doi.org/10.1016/j.jvoice.2025.04.019","url":null,"abstract":"<p><strong>Background: </strong>The vocal fold tissue modifications and related dysphonia caused by laryngopharyngeal reflux disease (LPRD) remain a controversial topic in laryngology. Investigation of human vocal fold tissue exposed to reflux content can provide valuable insights. This systematic review aimed to summarize the current knowledge about LPRD-induced human vocal fold tissue modifications to better understand LPRD pathophysiology and LPRD-related voice disorders.</p><p><strong>Methods: </strong>A PubMed, Embase, and Web of Science database search was carried out by two investigators for studies investigating human laryngeal mucosa injuries and histological modifications related to LPRD refluxate, and their potential mechanistic associations with voice quality impairments according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses statements.</p><p><strong>Results: </strong>Of 389 retrieved articles, 24 experimental studies met the inclusion criteria. Studies demonstrate that laryngeal, particularly vocal fold, biopsies of patients with suspected LPRD reveal a substantial number of histological and functional alterations, including inflammatory cell infiltration, cell junction proteolysis, intercellular space dilatation, pepsin-induced cell DNA damage, and increases in oxidative stress mediators and tissue injuries. Functional impairment of defensive mechanisms through downregulation of carbonic anhydrases (CA III) and protective mucins (MUC2, MUC4, and MUC5AC) can consist of favoring factor of tissue injuries. Emerging studies reported evidence of tissue remodeling through matrix metalloproteinase activation and metabolic alterations included increased Glut-1 and sphingosine pathway activation, potentially linking LPRD to leukoplakia development. No studies addressed the potential effects of elastase, bile salts, trypsin, and lipases in non-acidic (weakly acidic or alkaline) gaseous environment.</p><p><strong>Conclusion: </strong>This systematic review demonstrates that LPRD and pepsin induce cellular alterations in vocal fold and laryngeal tissues, highlighting potential pathogenic mechanisms and identifying biomarkers that may guide future diagnostic and therapeutic strategies.</p>","PeriodicalId":49954,"journal":{"name":"Journal of Voice","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144054954","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Clinical Findings, Feasibility, and Patient Tolerance of Nasopharyngeal Dx-pH System for Detecting Nasopharyngeal Reflux Disease.","authors":"Glenda Mantione, Gaetan Cavelier, Alexandra Rodriguez, Didier Dequanter, Sven Saussez, Stephane Hans, Antonino Maniaci, Giannicola Iannella, Giovanni Dapri, Luigi Vaira, Carlos M Chiesa-Estomba, Luigi La Via, Luigi Bonavina, Jérôme R Lechien","doi":"10.1016/j.jvoice.2025.04.023","DOIUrl":"https://doi.org/10.1016/j.jvoice.2025.04.023","url":null,"abstract":"<p><strong>Objective: </strong>To investigate the feasibility, patient tolerance, and clinical findings of the Dx-pH system for detecting nasopharyngeal reflux disease (NRD).</p><p><strong>Methods: </strong>Patients with idiopathic and chronic nasal complaints were recruited from the European Reflux Clinic between July 2022 and July 2024. Patients underwent 24-hour nasal Dx-pH system for detecting NRD. Reflux symptom score, sinonasal outcome tool-22, and reflux sign assessment were used to document symptoms and findings. A tolerance 19-item questionnaire was used to evaluate the symptom prevalence and severity of the probe placement and position throughout the 24-hour testing, ranging from 0 (no annoyance) to 95 (severe annoyance).</p><p><strong>Results: </strong>Twenty-three patients completed the evaluations (11 females). The mean age was 51.5 ± 17.0 years. Eighteen (78.3%) patients had NRD with a mean number of nasopharyngeal reflux events of 67.1 ± 65.9. Mulberry inferior turbinate was reported in 15 patients (65.2%), nasal dryness in nine patients (39.1%), and crusting in six patients (26.1%). The mean tolerance score was 15.2 ± 11.9. The most prevalent symptoms during the 24-hour pH-testing included nasal discomfort during probe placement (73.9%), throat discomfort during probe placement (69.6%), overall discomfort throughout the testing period (69.6%), cough during the testing day (65.2%), and postnasal drip sensation during the monitoring period (60.9%). Patients reported the highest discomfort scores for overall discomfort during the testing night and throat discomfort during probe placement. Significant positive correlations were observed between patient-reported tolerance difficulties and otolaryngological reflux symptom severity (r_s = 0.644, P = 0.002) and mulberry inferior turbinate (r_s = 0.432; P = 0.045).</p><p><strong>Conclusion: </strong>The Dx-pH system effectively detects NRD with acceptable patient tolerance, though discomfort correlates with symptom severity. This diagnostic approach suggests a high NRD prevalence among chronic nasal complaint patients.</p>","PeriodicalId":49954,"journal":{"name":"Journal of Voice","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143992522","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Initial Screening of Laryngeal Tumors via Voice Acoustic Analysis Based on Siamese Network Under Small Samples.","authors":"Zhenzhen You, Delong Sun, Zhenghao Shi, Shuangli Du, Xinhong Hei, Demin Kong, Xiaoying Du, Jing Yan, Xiaoyong Ren, Jin Hou","doi":"10.1016/j.jvoice.2025.03.043","DOIUrl":"https://doi.org/10.1016/j.jvoice.2025.03.043","url":null,"abstract":"<p><strong>Objective: </strong>The initial screening of laryngeal tumors via voice acoustic analysis is based on the clinician's experience that is subjective. This article introduces a Siamese network with an auxiliary gender classifier for automated, accurate, and objective initial screening of laryngeal tumors based on voice signals.</p><p><strong>Methods: </strong>The study involved 71 tumor patients and 293 non-tumor subjects of Chinese Mandarin. This dataset was divided into a training set and a test set in a ratio of 4:1. We applied nine data augmentation techniques to enlarge the voice training set and extracted the corresponding mel-frequency cepstral coefficients (MFCC) maps. The MFCC maps were randomly paired and fed into the proposed Siamese network to achieve multitask classification for tumor and non-tumor, woman and man. The performance of the proposed model was compared with one machine learning method and six classical deep learning models with and without the auxiliary gender classifier.</p><p><strong>Results: </strong>Experiments demonstrate the superiority of the proposed network compared with the reference models. The proposed model achieved an overall accuracy of 0.9437, an F score of 0.8462, a precision of 0.9167, a sensitivity of 0.7857, and a specificity of 0.9825.</p><p><strong>Conclusion: </strong>The proposed network can assist in the initial screening of laryngeal tumors through voice acoustic analysis. The initial screening solely through voice acoustic analysis can help individuals seek medical assistance outside the hospitals and reduce the burden on doctors as well.</p>","PeriodicalId":49954,"journal":{"name":"Journal of Voice","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143993786","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparison of Voice Stimulability Testing in Telehealth and In-Person Formats.","authors":"Melanie A Turner, Laura E Toles","doi":"10.1016/j.jvoice.2025.04.016","DOIUrl":"https://doi.org/10.1016/j.jvoice.2025.04.016","url":null,"abstract":"<p><strong>Objective: </strong>Telehealth is a useful platform for behavioral voice evaluation and therapy and has been shown to be noninferior to in-person services. However, in the existing literature, therapy outcomes are only compared after a complete course of voice therapy. Voice stimulability testing is an important part of the behavioral voice evaluation and can guide treatment decisions, estimate prognosis for improvement, and facilitate the patient's readiness for change. It is unclear whether immediate changes that occur during stimulability testing differ between in-person and telehealth delivery models. This study examines whether stimulability ratings differ significantly between in-person and telehealth voice evaluations.</p><p><strong>Methods: </strong>Charts from 86 patients who underwent a voice evaluation via telehealth and 86 patients who participated in an in-person evaluation were reviewed. Telehealth and in-person patients were matched based on voice diagnosis (structural, functional, and neurological) and which clinician they saw. Chart review included documentation of age, diagnosis, clinician, degree of overall stimulability (low, medium, and high), and degree of stimulability to specific tasks. Multinomial logistic regression was completed to determine whether there were differences in stimulability levels between practice modalities when controlling for age.</p><p><strong>Results: </strong>The overall logistic regression model was not statistically significant (P = 0.091). There were no statistically significant differences in overall stimulability level between patients undergoing in-person evaluation and those who participated in telehealth evaluations, even when controlling for age.</p><p><strong>Conclusions: </strong>Voice stimulability testing can impact a clinician's therapeutic and prognostic decisions, so it is important that stimulability can be judged consistently across all treatment settings. The results from this study found that estimates of stimulability did not differ between telehealth and in-person modalities. These findings provide further support for the use of telehealth as a noninferior modality for voice therapy services.</p>","PeriodicalId":49954,"journal":{"name":"Journal of Voice","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144056505","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Preliminary Effects of Respiratory Lung Volume Training (RLVT) on Hyolaryngeal Kinematics in People With Primary Muscle Tension Dysphonia.","authors":"Lisandra Pereyra Maldonado, Anya Freedman-Doan, Soren Y Lowell","doi":"10.1016/j.jvoice.2025.03.047","DOIUrl":"https://doi.org/10.1016/j.jvoice.2025.03.047","url":null,"abstract":"<p><strong>Objectives: </strong>The purpose of this preliminary study was to determine the effects of a novel respiratory-based intervention on hyolaryngeal kinematics in people with primary muscle tension dysphonia (pMTD) using ultrasonography and acoustic measures.</p><p><strong>Method: </strong>Pretraining to post-training changes in hyolaryngeal kinematics were studied for 11 participants with pMTD using ultrasonography to objectively measure and compare the relative positions of the hyoid bone and thyroid cartilage before and after treatment. Displacement of the hyoid bone and thyroid cartilage was measured during sustained phonation of the vowel /ɑ/, before and after completion of Respiratory Lung Volume Training (RLVT), which trains the use of higher speech lung volume initiation and termination levels while implementing real-time visual biofeedback. Hyolaryngeal position was determined from still frames extracted from ultrasound video recordings, with measures normalized to reflect change from rest during voicing for each participant. The associations between hyolaryngeal kinematics and acoustic measures of voice were also investigated.</p><p><strong>Results: </strong>Thyroid-to-hyoid displacement during voicing was significantly lower after completion of RLVT. There were no differences in hyoid-to-mandible displacement during voicing when comparing pretraining to post-training timepoints. Moderate-to-strong associations between thyroid-to-hyoid displacement and acoustic voice measures were identified for one or more speaking contexts.</p><p><strong>Conclusions: </strong>Laryngeal elevation during voicing appears to be reduced after completion of RLVT in people with pMTD. Mechanical and functional interactions of the respiratory and laryngeal subsystems likely contributed to this kinematic change and the associations between laryngeal kinematics and acoustic voice features. Ultrasonography-based anatomic displacement measurements may provide an objective and sensitive method for determining treatment-related change in hyolaryngeal kinematics.</p>","PeriodicalId":49954,"journal":{"name":"Journal of Voice","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144041702","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Prevalence of Dysphonia in Metal Singers and the Impact of Vocal Education: A Subjective Analysis","authors":"Carmen Unterhofer ,&nbsp;Judith Marscheider ,&nbsp;Sophie Auer,&nbsp;Olivia Jeleff-Wölfler,&nbsp;Simone Graf","doi":"10.1016/j.jvoice.2022.12.004","DOIUrl":"10.1016/j.jvoice.2022.12.004","url":null,"abstract":"<div><div>On average, the prevalence of subjectively indicated dysphonia in singers is higher than in the general population. When using extreme vocal techniques supraglottic structures are involved to protect the vocal folds. Most of the metal singers learn this specific singing style autodidactically. Vocal training can have a positive effect on the voice and prevent voice problems. The aim of the study was to find out the prevalence of dysphonia in metal singers and the impact this has on the profession or the vocal education.</div><div>The subjects had to fill out an online questionnaire including two validated and standardized questionnaires about the speaking and singing voice: The short version of the Voice Handicap Index (VHI 12) and the Singing Voice Handicap Index (SVHI).</div><div>The prevalence of dysphonia in metal singers is 12.2%. 9 out of 74 participants showed an increased (S)VHI score. 6 of these participants learned singing autodidactically. No dysphonia was found in any singer who sings full-time or studied singing. The correlation between education or profession and dysphonia was statistically not significant. More women than men studied or took singing lessons. This difference was statistically significant.</div><div>The prevalence of 12.2% in our study is quite low and contrary to our assumption. We can assume, that a vocal education for the metal singing style is useful but not mandatory and the number of persons with dysphonia through metal singing is low. A healthy vocal technique does not necessarily have to be the result of voice lessons.</div></div>","PeriodicalId":49954,"journal":{"name":"Journal of Voice","volume":"39 3","pages":"Pages 853.e21-853.e31"},"PeriodicalIF":2.5,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9100831","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effects of Virtual Teaching on Swedish Teachers’ Voices During the COVID-19 Pandemic","authors":"Paul M. Evitts ,&nbsp;Maria Allebeck ,&nbsp;Olivia Esmerelda Åberg","doi":"10.1016/j.jvoice.2022.12.022","DOIUrl":"10.1016/j.jvoice.2022.12.022","url":null,"abstract":"<div><h3>Purpose</h3><div>Since the beginning of 2020, teachers all over the world have had to switch over to virtual teaching because of the COVID-19 pandemic. This change of teaching mode has unknown consequences on the vocal health of teachers who already experience voice disorders at a rate nearly double that of the general population. Therefore, the purpose of this study is to investigate teachers’ self-reported voice problems during virtual teaching compared to face-to-face teaching and how they are associated to perceived risk factors in the work environment.</div></div><div><h3>Method</h3><div>Participants included 141 teachers (49 males, 92 females) in high schools and higher education in Sweden who responded to an internet-based survey. Information was collected about participants’ self-reported voice handicap using a translated version of the Voice Handicap Index-10 (VHI-10). Information about self-reported VTD (Vocal Tract Discomfort) symptoms and dysphonia was collected in regard to both modes of teaching (virtual vs face-to- face). The survey also included questions on risk factors related to vocal health.</div></div><div><h3>Results</h3><div>Self-reported voice handicap and VTD symptoms were slightly lower during periods of virtual teaching compared to periods of face-to-face teaching. There was a lower frequency of dysphonia symptoms during virtual teaching compared to face-to- face teaching, however the difference was not statistically significant. In addition, 34% of teachers reported experiencing more voice problems when teaching face-to-face while 15% reported more voice problems when teaching virtually. The most reported VTD symptoms during both virtual and face-to-face teaching were having a dry and a tight throat. The dysphonia symptoms with the highest reported frequency were a tense voice and hoarseness in both modes of teaching. Risk factors associated with higher prevalence of voice symptoms and/or higher levels of voice handicap during virtual teaching were air quality and straining the voice while teaching face-to-face. In addition, those teachers who reported more voice problems while teaching virtually also reported feeling more stressed.</div></div><div><h3>Conclusion</h3><div>The results showed a slight decrease in voice symptoms and voice handicap during virtual teaching compared to face-to-face teaching. Although multifactorial, results suggest that a potential positive effect may be attributed to better air quality in the work environment and more favorable acoustic conditions preventing teachers from straining their voice.</div></div>","PeriodicalId":49954,"journal":{"name":"Journal of Voice","volume":"39 3","pages":"Pages 843.e9-843.e21"},"PeriodicalIF":2.5,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9182093","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Use of Laryngeal Electromyography to Determine Surgical Success Following Anastomosis and Implantation Reinnervation Surgeries for Vocal Fold Paralysis: A Systematic Review","authors":"Camryn Marshall ,&nbsp;Divya Balchander ,&nbsp;Ghiath Alnouri ,&nbsp;Robert T. Sataloff","doi":"10.1016/j.jvoice.2022.11.030","DOIUrl":"10.1016/j.jvoice.2022.11.030","url":null,"abstract":"<div><h3>Background</h3><div><span>The goal of this systematic review is to elucidate an evidence-based review of the benefit of laryngeal </span>electromyography<span> (LEMG) in determining surgical success following reanastamosis and implantation reinnervation<span> surgery for treating vocal fold paralysis.</span></span></div></div><div><h3>Methods</h3><div>We searched the following databases: PubMed, Embase and MEDLINE from inception to September 2022. Articles that included postoperative LEMG as a tool to evaluate patients after surgery for vocal fold paralysis were included.</div></div><div><h3>Results</h3><div>A total of 203 studies were identified, 106 were read in full, and 11 that met the inclusion criteria were systematically reviewed.</div></div><div><h3>Conclusion</h3><div>The available evidence suggests that postoperative LEMG data is consistent with visual, perceptual, and acoustic analysis<span> and therefore is a reliable tool in determining surgical success. LEMG provides both qualitative and quantitative evaluation, rather than the qualitative evaluation provided by standard methods, which yields additional parameters that more effectively and definitively characterize percent of muscle reinnervation at each stage and timeline the patient's recovery.</span></div></div>","PeriodicalId":49954,"journal":{"name":"Journal of Voice","volume":"39 3","pages":"Pages 775-786"},"PeriodicalIF":2.5,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10346096","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Additive Effectiveness of Inspiratory Muscle Training on Glottic Closure and Subjective Voice Outcomes of Patients With Benign Lesion After Hyaluronic Acid Laryngoplasty","authors":"Hsin-Hao Liou ,&nbsp;Miyuki Hsing-Chun Hsieh ,&nbsp;David Shang-Yu Hung ,&nbsp;Hui-Ling Liu ,&nbsp;Ishan Lee ,&nbsp;Yi-Chih Lin ,&nbsp;Jenn-Ren Hsiao ,&nbsp;Cheng-Chih Huang ,&nbsp;Chun-Yen Ou ,&nbsp;Chan-Chi Chang ,&nbsp;Wei-Ting Lee ,&nbsp;Sen-Tien Tsai ,&nbsp;Shu-Wei Tsai","doi":"10.1016/j.jvoice.2022.10.021","DOIUrl":"10.1016/j.jvoice.2022.10.021","url":null,"abstract":"<div><h3>Objectives</h3><div><span><span>For patients with glottic insufficiency disease, injection laryngoplasty is a rapid and efficient management option that complements voice therapy. Some studies have indicated that </span>respiratory muscle training may also show promise in patients with voice disorders. However, the effect of respiratory muscle training in patients with glottic insufficiency was reported to be limited, and whether it provides additional benefit after standard management requires further evaluation. We aimed to investigate the effectiveness of inspiratory muscle training on </span>glottis<span> closure and patient-reported voice quality in glottic insufficiency patients who had been treated with hyaluronic acid injection.</span></div></div><div><h3>Study Design</h3><div>Retrospective observational study.</div></div><div><h3>Methods</h3><div>We included 46 patients with glottic insufficiency who had undergone hyaluronic acid injection. Twenty of them had undergone inspiratory muscle training during three months. We measured patients’ changes in glottic status according to the normalized glottal gap area and bowing index, as well as voice quality of life according to the voice handicap index 10 and the voice outcome survey, before and after training.</div></div><div><h3>Results</h3><div>Patients who underwent inspiratory muscle training had higher odds of experiencing better improvement in all scores. The range of odds ratios ranged from 2.5 to 6.3 for changes in scores, and from 3.8 to 22.2 for changes in score percentages. Of note, the effect of training on percentage changes in the normalized glottal gap area score was significant (<em>P</em><span>= 0.0127) after adjustment for the duration of vocal disease, body mass index<span> and BMI, and history of gastroesophageal reflux disease.</span></span></div></div><div><h3>Conclusions</h3><div>Inspiratory muscle training can improve the glottal gap after injection laryngoplasty, and may be applied in clinical practice.</div></div>","PeriodicalId":49954,"journal":{"name":"Journal of Voice","volume":"39 3","pages":"Pages 728-735"},"PeriodicalIF":2.5,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10399737","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Different Performances of Machine Learning Models to Classify Dysphonic and Non-Dysphonic Voices","authors":"Danilo Rangel Arruda Leite ,&nbsp;Ronei Marcos de Moraes ,&nbsp;Leonardo Wanderley Lopes","doi":"10.1016/j.jvoice.2022.11.001","DOIUrl":"10.1016/j.jvoice.2022.11.001","url":null,"abstract":"<div><h3>Objective</h3><div>To analyze the performance of 10 different machine learning (ML) classifiers for discrimination between dysphonic and non-dysphonic voices, using a variance threshold as a method for the selection and reduction of acoustic measurements used in the classifier.</div></div><div><h3>Method</h3><div>We analyzed 435 samples of individuals (337 female and 98 male), with a mean age of 41.07 ± 13.73 years, of which 384 were dysphonic and 51 were non-dysphonic. From the sustained /ε/ vowel sample, 34 acoustic measurements were extracted, including traditional perturbation and noise measurements, cepstral/spectral measurements, and measurements based on nonlinear models<span>. The variance method was used to select the best set of acoustic measurements. We tested the performance of the best-selected set with 10 ML classifiers using precision, sensitivity, specificity, accuracy, and F1-Score measurements. The kappa coefficient was used to verify the reproducibility between the two datasets (training and testing).</span></div></div><div><h3>Results</h3><div><span>The naive Bayes (NB) and stochastic gradient descent classifier (SGDC) models performed best in terms of accuracy, AUC, sensitivity, and specificity for a reduced </span><em>dataset</em> of 15 acoustic measures compared to the full <em>dataset</em> of 34 acoustic measures. SGDC and NB obtained the best performance results, with an accuracy of 0.91 and 0.76, respectively. These two classifiers presented moderate agreement, with a Kappa of 0.57 (SGDC) and 0.45 (NB).</div></div><div><h3>Conclusion</h3><div>Among the tested models, the NB and SGDC models performed better in discriminating between dysphonic and non-dysphonic voices from a set of 15 acoustic measures.</div></div>","PeriodicalId":49954,"journal":{"name":"Journal of Voice","volume":"39 3","pages":"Pages 577-590"},"PeriodicalIF":2.5,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10399739","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}