LarynxFormer: a transformer-based framework for processing and segmenting laryngeal images

Manual diagnostic methods for assessing exercise-induced laryngeal obstruction (EILO) contain human bias and can lead to subjective decisions. Several studies have proposed machine learning methods for segmenting laryngeal structures to automate and make diagnostic outcomes more objective. Four stat...

Full description

Saved in:
Bibliographic Details
Main Authors: Rune Mæstad, Abdul Hanan, Haakon Kristian Kvidaland, Hege Clemm, Reza Arghandeh
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-07-01
Series:Frontiers in Digital Health
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fdgth.2025.1459136/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849429207144202240
author Rune Mæstad
Abdul Hanan
Haakon Kristian Kvidaland
Haakon Kristian Kvidaland
Hege Clemm
Hege Clemm
Reza Arghandeh
author_facet Rune Mæstad
Abdul Hanan
Haakon Kristian Kvidaland
Haakon Kristian Kvidaland
Hege Clemm
Hege Clemm
Reza Arghandeh
author_sort Rune Mæstad
collection DOAJ
description Manual diagnostic methods for assessing exercise-induced laryngeal obstruction (EILO) contain human bias and can lead to subjective decisions. Several studies have proposed machine learning methods for segmenting laryngeal structures to automate and make diagnostic outcomes more objective. Four state-of-the-art models for laryngeal image segmentation are implemented, trained, and compared using our pre-processed dataset containing laryngeal images derived from continuous laryngoscopy exercise-test (CLE-test) data. These models include both convolutional-based and transformer-based methods. We propose a new framework called LarynxFormer, consisting of a pre-processing pipeline, transformer-based segmentation, and post-processing of laryngeal images. This study contributes to the investigation of using machine learning as a diagnostic tool for EILO. Furthermore, we show that a transformer-based approach for larynx segmentation outperforms conventional state-of-the-art image segmentation methods in terms of performance metrics and computational speed, demonstrating up to 2x faster inference time compared to the other methods.
format Article
id doaj-art-35cba6921e974cd1b38fc7935dd2db90
institution Kabale University
issn 2673-253X
language English
publishDate 2025-07-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Digital Health
spelling doaj-art-35cba6921e974cd1b38fc7935dd2db902025-08-20T03:28:26ZengFrontiers Media S.A.Frontiers in Digital Health2673-253X2025-07-01710.3389/fdgth.2025.14591361459136LarynxFormer: a transformer-based framework for processing and segmenting laryngeal imagesRune Mæstad0Abdul Hanan1Haakon Kristian Kvidaland2Haakon Kristian Kvidaland3Hege Clemm4Hege Clemm5Reza Arghandeh6Faculty of Engineering and Science, Western Norway University of Applied Sciences, Bergen, Vestland, NorwayFaculty of Engineering and Science, Western Norway University of Applied Sciences, Bergen, Vestland, NorwayFaculty of Health and Social Sciences, Western Norway University of Applied Sciences, Bergen, Vestland, NorwayDepartment of Pediatric and Adolescent Medicine, Haukeland University Hospital, Bergen, Vestland, NorwayDepartment of Clinical Science, University of Bergen, Bergen, Vestland, NorwayDepartment of Head and Neck, Haukeland University Hospital, Bergen, Vestland, NorwayFaculty of Engineering and Science, Western Norway University of Applied Sciences, Bergen, Vestland, NorwayManual diagnostic methods for assessing exercise-induced laryngeal obstruction (EILO) contain human bias and can lead to subjective decisions. Several studies have proposed machine learning methods for segmenting laryngeal structures to automate and make diagnostic outcomes more objective. Four state-of-the-art models for laryngeal image segmentation are implemented, trained, and compared using our pre-processed dataset containing laryngeal images derived from continuous laryngoscopy exercise-test (CLE-test) data. These models include both convolutional-based and transformer-based methods. We propose a new framework called LarynxFormer, consisting of a pre-processing pipeline, transformer-based segmentation, and post-processing of laryngeal images. This study contributes to the investigation of using machine learning as a diagnostic tool for EILO. Furthermore, we show that a transformer-based approach for larynx segmentation outperforms conventional state-of-the-art image segmentation methods in terms of performance metrics and computational speed, demonstrating up to 2x faster inference time compared to the other methods.https://www.frontiersin.org/articles/10.3389/fdgth.2025.1459136/fullexercise-induced laryngeal obstructioncontinuous laryngoscopy exercise testmachine learningartificial intelligenceimage segmentation
spellingShingle Rune Mæstad
Abdul Hanan
Haakon Kristian Kvidaland
Haakon Kristian Kvidaland
Hege Clemm
Hege Clemm
Reza Arghandeh
LarynxFormer: a transformer-based framework for processing and segmenting laryngeal images
Frontiers in Digital Health
exercise-induced laryngeal obstruction
continuous laryngoscopy exercise test
machine learning
artificial intelligence
image segmentation
title LarynxFormer: a transformer-based framework for processing and segmenting laryngeal images
title_full LarynxFormer: a transformer-based framework for processing and segmenting laryngeal images
title_fullStr LarynxFormer: a transformer-based framework for processing and segmenting laryngeal images
title_full_unstemmed LarynxFormer: a transformer-based framework for processing and segmenting laryngeal images
title_short LarynxFormer: a transformer-based framework for processing and segmenting laryngeal images
title_sort larynxformer a transformer based framework for processing and segmenting laryngeal images
topic exercise-induced laryngeal obstruction
continuous laryngoscopy exercise test
machine learning
artificial intelligence
image segmentation
url https://www.frontiersin.org/articles/10.3389/fdgth.2025.1459136/full
work_keys_str_mv AT runemæstad larynxformeratransformerbasedframeworkforprocessingandsegmentinglaryngealimages
AT abdulhanan larynxformeratransformerbasedframeworkforprocessingandsegmentinglaryngealimages
AT haakonkristiankvidaland larynxformeratransformerbasedframeworkforprocessingandsegmentinglaryngealimages
AT haakonkristiankvidaland larynxformeratransformerbasedframeworkforprocessingandsegmentinglaryngealimages
AT hegeclemm larynxformeratransformerbasedframeworkforprocessingandsegmentinglaryngealimages
AT hegeclemm larynxformeratransformerbasedframeworkforprocessingandsegmentinglaryngealimages
AT rezaarghandeh larynxformeratransformerbasedframeworkforprocessingandsegmentinglaryngealimages