A Deep‐Learning Model for Multi‐class Audio Classification of Vocal Fold Pathologies in Office Stroboscopy
The Laryngoscope · 2025
DOI: 10.1002/lary.32036 ↗Kurz gefasst
Validierungsstudie in der Zeitschrift The Laryngoscope, 2025. Verfasst von Yeo E Kim und 12 weiteren Autor:innen. Beteiligte Einrichtung: Sean Parker Institute for the Voice, Department of Otolaryngology-Head and Neck Surgery, Weill Cornell Medicine, New York, New York, U.S.A. Hauptthemen laut MEDLINE: Deep Learning, Vocal Cords, Stroboscopy, Vocal Cord Paralysis. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIGMS NIH HHS, NIA NIH HHS, ODCDC CDC HHS.
Abstract (Originalsprache)
Objective
To develop and validate a deep‐learning classifier trained on voice data extracted from videolaryngostroboscopy recordings, differentiating between three different vocal fold (VF) states: healthy (HVF), unilateral paralysis (UVFP), and VF lesions, including benign and malignant pathologies.
Methods
Patients with UVFP ( n = 105), VF lesions ( n = 63), and HVF ( n = 41) were retrospectively identified. Voice samples were extracted from stroboscopic videos (Pentax Laryngeal Strobe Model 9400), including sustained /i/ phonation, pitch glide, and /i/ sniff task. Extracted audio files were converted into Mel‐spectrograms. Voice samples were independently divided into training (80%), validation (10%), and test (10%) by patient. Pretrained ResNet18 models were trained to classify (1) HVF and pathological VF (lesions and UVFP), and (2) HVF, UVFP, and VF lesions. Both classifiers were further validated on an external dataset consisting of 12 UVFP, 13 VF lesions, and 15 HVF patients. Model performances were evaluated by accuracy and F1‐score.
Results
When evaluated on a hold‐out test set, the binary classifier demonstrated stronger performance compared to the multi‐class classifier (accuracy 83% vs. 40%; F1‐score 0.90 vs. 0.36). When evaluated on an external dataset, the binary classifier achieved an accuracy of 63% and F1‐score of 0.48, compared to 35% and 0.25 for the multi‐class classifier.
Conclusions
Deep‐learning classifiers differentiating HVF, UVFP, and VF lesions were developed using voice data from stroboscopic videos. Although healthy and pathological voice were differentiated with moderate accuracy, multi‐class classification lowered model performance. The model performed poorly on an external dataset. Voice captured in stroboscopic videos may have limited diagnostic value, though further studies are needed. Level of Evidence 4 Laryngoscope , 135:2428–2436, 2025
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Sean Parker Institute for the Voice, Department of Otolaryngology-Head and Neck Surgery, Weill Cornell Medicine, New York, New York, U.S.A.
- Cornell Tech, New York, New York, U.S.A.
- The Center for Voice and Swallowing, Department of Otolaryngology-Head and Neck Surgery, New York-Presbyterian Hospital, Columbia University Irvine Medical Center, New York City, New York, U.S.A.
- Weill Cornell Medicine-Qatar, Doha, Qatar.
- Englander Institute for Precision Medicine, Weill Cornell Medicine, New York, New York, U.S.A.
- Department of Radiology, Weill Cornell Medicine, New York, New York, U.S.A.
ORCID-Profile: Yeo E Kim, Christine Colasacco, Michael Pitman, Lucian Sulica, Anaïs Rameau
Förderung
- NIGMS NIH HHS
- NIA NIH HHS
- ODCDC CDC HHS
- NIH HHS
- National Institute on Aging
- Common Fund
Zitieren
APA
Kim, Y. E., Dobko, M., Li, H., Shao, T., Periyakoil, P., Tipton, C., Colasacco, C., Serpedin, A., Elemento, O., Sabuncu, M., Pitman, M., Sulica, L., & Rameau, A. (2025). A Deep‐Learning Model for Multi‐class Audio Classification of Vocal Fold Pathologies in Office Stroboscopy. The Laryngoscope. https://doi.org/10.1002/lary.32036
Vancouver
Kim YE, Dobko M, Li H, Shao T, Periyakoil P, Tipton C, et al. A Deep‐Learning Model for Multi‐class Audio Classification of Vocal Fold Pathologies in Office Stroboscopy. The Laryngoscope. 2025. doi:10.1002/lary.32036
BibTeX
@article{Kim2025,
author = {Kim, Yeo E and Dobko, Maria and Li, Haomiao and Shao, Tianlan and Periyakoil, Preethi and Tipton, Courtney and Colasacco, Christine and Serpedin, Aisha and Elemento, Olivier and Sabuncu, Mert and Pitman, Michael and Sulica, Lucian and Rameau, Anaïs},
title = {A Deep‐Learning Model for Multi‐class Audio Classification of Vocal Fold Pathologies in Office Stroboscopy},
journal = {The Laryngoscope},
year = {2025},
doi = {10.1002/lary.32036}
}
EUVIMED