Linear versus nonlinear modeling of dysphonia severity: A comparison between Acoustic Voice Quality Index and machine learning
Journal of Communication Disorders · 2026-05 · 121 · 106648
DOI: 10.1016/j.jcomdis.2026.106648 ↗Kurz gefasst
Vergleichsstudie in der Zeitschrift Journal of Communication Disorders, 2026-05. Verfasst von Ahmed M Yousef und 5 weiteren Autor:innen. Beteiligte Einrichtung: Center for Laryngeal Surgery and Voice Rehabilitation, Massachusetts General Hospital, One Bowdoin Square, 11th Floor, Boston, MA 02114, USA; Departm…. Hauptthemen laut MEDLINE: Machine Learning, Dysphonia, Voice Quality, Speech Acoustics. Die Arbeit nennt 79 Literaturangaben. Gefördert von: National Institutes of Health, National Institute on Deafness and Other Communication Disorders.
Abstract (Originalsprache)
PURPOSE: This study compares the Acoustic Voice Quality Index (AVQI-3) with machine learning (ML) models to evaluate their clinical utility for estimating voice quality. METHODS: Audio from 187 American English speakers (49 healthy, 138 with voice disorders) was rated for overall voice quality by six voice specialists. AVQI-3 and its six acoustic parameters were extracted, and these parameters were used to train 14 ML models (linear, curvilinear, nonlinear). Correlations and classification accuracy (normal-mild vs. moderate-severe) were compared against perceptual ratings as ground truth. RESULTS: AVQI-3 correlated strongly with perceptual ratings (Spearman rs=0.75), comparable to linear and curvilinear models, yet outperformed most nonlinear models. At a cutoff score of 2.52, AVQI-3 achieved the highest classification accuracy (0.92) with balanced sensitivity (0.90) and specificity (0.93). Among ML models, linear regression performed best (rs=0.77, accuracy=0.92, sensitivity=1.0, specificity=0.89), whereas nonlinear models showed reduced performance (average rs=0.74, accuracy=0.87, sensitivity=0.95, specificity=0.85). CONCLUSION: AVQI-3 is a simple, accessible index that quantifies voice quality as effectively as complex ML models. This is supported by the best-performing ML models being linear, indicating that linear combinations of acoustic measures are effective, accurate, and clinically interpretable, whereas added nonlinear complexity offered little benefit.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Center for Laryngeal Surgery and Voice Rehabilitation, Massachusetts General Hospital, One Bowdoin Square, 11th Floor, Boston, MA 02114, USA; Department of Surgery, Harvard Medical School, Boston, MA 02115, USA; Departm…
- Department of Communication Sciences and Disorders, University of Iowa, Iowa City, IA 52242, USA; Department of Communicative Sciences and Disorders, Michigan State University, East Lansing, MI 48824, USA.
- Department of Communication Sciences and Disorders, University of Iowa, Iowa City, IA 52242, USA.
- Lakeshore Professional Voice Center, Lakeshore Ear, Nose & Throat Center, St. Clair Shores, MI 48081, USA.
Förderung
- National Institutes of Health
- National Institute on Deafness and Other Communication Disorders
Zitieren
APA
Yousef, A. M., Castillo-Allendes, A., Berardi, M. L., Codino, J., Rubin, A. D., & Hunter, E. J. (2026). Linear versus nonlinear modeling of dysphonia severity: A comparison between Acoustic Voice Quality Index and machine learning. Journal of Communication Disorders, 121, 106648. https://doi.org/10.1016/j.jcomdis.2026.106648
Vancouver
Yousef AM, Castillo-Allendes A, Berardi ML, Codino J, Rubin AD, Hunter EJ. Linear versus nonlinear modeling of dysphonia severity: A comparison between Acoustic Voice Quality Index and machine learning. Journal of Communication Disorders. 2026;121:106648. doi:10.1016/j.jcomdis.2026.106648
BibTeX
@article{Yousef2026,
author = {Yousef, Ahmed M and Castillo-Allendes, Adrián and Berardi, Mark L and Codino, Juliana and Rubin, Adam D and Hunter, Eric J},
title = {Linear versus nonlinear modeling of dysphonia severity: A comparison between Acoustic Voice Quality Index and machine learning},
journal = {Journal of Communication Disorders},
year = {2026},
volume = {121},
pages = {106648},
doi = {10.1016/j.jcomdis.2026.106648}
}
EUVIMED