Diagnosis of unilateral vocal fold paralysis using auto-diagnostic deep learning model

Abstract Unilateral vocal fold paralysis (UVFP) is a condition characterized by impaired vocal fold mobility, typically diagnosed using laryngeal videoendoscopy. While deep learning (DL) models using static images have been explored for UVFP detection, they often lack the ability to assess vocal fol...

Full description

Saved in:
Bibliographic Details
Main Authors: Kyoung Ok Yang, So Young Kim, Chang Won Kang, Jeong Seon Choi, Yong Bae Ji, Kyung Tae, Jun Won Choi, Chang Myeon Song
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-09797-z
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849332869280825344
author Kyoung Ok Yang
So Young Kim
Chang Won Kang
Jeong Seon Choi
Yong Bae Ji
Kyung Tae
Jun Won Choi
Chang Myeon Song
author_facet Kyoung Ok Yang
So Young Kim
Chang Won Kang
Jeong Seon Choi
Yong Bae Ji
Kyung Tae
Jun Won Choi
Chang Myeon Song
author_sort Kyoung Ok Yang
collection DOAJ
description Abstract Unilateral vocal fold paralysis (UVFP) is a condition characterized by impaired vocal fold mobility, typically diagnosed using laryngeal videoendoscopy. While deep learning (DL) models using static images have been explored for UVFP detection, they often lack the ability to assess vocal fold dynamics. We developed an auto-diagnostic DL system for UVFP using both image-based and video-based models. Using laryngeal videoendoscopic data from 500 participants, the model was trained and validated on 2639 video clips. The image-based DL model achieved over 98% accuracy for UVFP detection, but demonstrated limited performance in predicting laterality and paralysis type. In contrast, the video-based model achieved comparable accuracy (about 99%) in detecting UVFP, and substantially higher accuracy in predicting laterality and paralysis type, outperforming the image-based model in overall diagnostic utility. These results demonstrate the advantages of incorporating temporal motion cues in video-based analysis and support the use of DL for comprehensive, multi-task assessment of UVFP. This automated approach demonstrates high diagnostic performance and may serve as a complementary tool to assist clinicians in the assessment of UVFP, particularly in enhancing workflow efficiency and supporting multi-dimensional interpretation of laryngeal motion.
format Article
id doaj-art-1b3a3eb2e726421d98d163c6aabd9682
institution Kabale University
issn 2045-2322
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-1b3a3eb2e726421d98d163c6aabd96822025-08-20T03:46:04ZengNature PortfolioScientific Reports2045-23222025-07-0115111010.1038/s41598-025-09797-zDiagnosis of unilateral vocal fold paralysis using auto-diagnostic deep learning modelKyoung Ok Yang0So Young Kim1Chang Won Kang2Jeong Seon Choi3Yong Bae Ji4Kyung Tae5Jun Won Choi6Chang Myeon Song7Department of Artificial Intelligence, Hanyang UniversityDepartment of Anatomy and Cell Biology, College of Medicine, Seoul National UniversityDepartment of Artificial Intelligence, Hanyang UniversityDepartment of Artificial Intelligence, Hanyang UniversityDepartment of Otolaryngology-Head and Neck Surgery, Hanyang University College of MedicineDepartment of Otolaryngology-Head and Neck Surgery, Hanyang University College of MedicineDepartment of Electrical and Computer Engineering, Seoul National UniversityDepartment of Otolaryngology-Head and Neck Surgery, Hanyang University College of MedicineAbstract Unilateral vocal fold paralysis (UVFP) is a condition characterized by impaired vocal fold mobility, typically diagnosed using laryngeal videoendoscopy. While deep learning (DL) models using static images have been explored for UVFP detection, they often lack the ability to assess vocal fold dynamics. We developed an auto-diagnostic DL system for UVFP using both image-based and video-based models. Using laryngeal videoendoscopic data from 500 participants, the model was trained and validated on 2639 video clips. The image-based DL model achieved over 98% accuracy for UVFP detection, but demonstrated limited performance in predicting laterality and paralysis type. In contrast, the video-based model achieved comparable accuracy (about 99%) in detecting UVFP, and substantially higher accuracy in predicting laterality and paralysis type, outperforming the image-based model in overall diagnostic utility. These results demonstrate the advantages of incorporating temporal motion cues in video-based analysis and support the use of DL for comprehensive, multi-task assessment of UVFP. This automated approach demonstrates high diagnostic performance and may serve as a complementary tool to assist clinicians in the assessment of UVFP, particularly in enhancing workflow efficiency and supporting multi-dimensional interpretation of laryngeal motion.https://doi.org/10.1038/s41598-025-09797-zVocal fold paralysisDeep learning modelAuto-diagnosisLaryngeal videoendoscopyMulti-task learningVocal fold movement
spellingShingle Kyoung Ok Yang
So Young Kim
Chang Won Kang
Jeong Seon Choi
Yong Bae Ji
Kyung Tae
Jun Won Choi
Chang Myeon Song
Diagnosis of unilateral vocal fold paralysis using auto-diagnostic deep learning model
Scientific Reports
Vocal fold paralysis
Deep learning model
Auto-diagnosis
Laryngeal videoendoscopy
Multi-task learning
Vocal fold movement
title Diagnosis of unilateral vocal fold paralysis using auto-diagnostic deep learning model
title_full Diagnosis of unilateral vocal fold paralysis using auto-diagnostic deep learning model
title_fullStr Diagnosis of unilateral vocal fold paralysis using auto-diagnostic deep learning model
title_full_unstemmed Diagnosis of unilateral vocal fold paralysis using auto-diagnostic deep learning model
title_short Diagnosis of unilateral vocal fold paralysis using auto-diagnostic deep learning model
title_sort diagnosis of unilateral vocal fold paralysis using auto diagnostic deep learning model
topic Vocal fold paralysis
Deep learning model
Auto-diagnosis
Laryngeal videoendoscopy
Multi-task learning
Vocal fold movement
url https://doi.org/10.1038/s41598-025-09797-z
work_keys_str_mv AT kyoungokyang diagnosisofunilateralvocalfoldparalysisusingautodiagnosticdeeplearningmodel
AT soyoungkim diagnosisofunilateralvocalfoldparalysisusingautodiagnosticdeeplearningmodel
AT changwonkang diagnosisofunilateralvocalfoldparalysisusingautodiagnosticdeeplearningmodel
AT jeongseonchoi diagnosisofunilateralvocalfoldparalysisusingautodiagnosticdeeplearningmodel
AT yongbaeji diagnosisofunilateralvocalfoldparalysisusingautodiagnosticdeeplearningmodel
AT kyungtae diagnosisofunilateralvocalfoldparalysisusingautodiagnosticdeeplearningmodel
AT junwonchoi diagnosisofunilateralvocalfoldparalysisusingautodiagnosticdeeplearningmodel
AT changmyeonsong diagnosisofunilateralvocalfoldparalysisusingautodiagnosticdeeplearningmodel