Localization and quantification of glottal gaps on deep learning segmentation of vocal folds

Abstract The entire glottis has mostly been the focus in the tracking of the vocal folds, both manually and automatically. From a treatment point of view, the various regions of the glottis are of specific interest. The aim of the study was to test if it was possible to supplement an existing convol...

Full description

Bibliographic Details
Main Authors: Mette Pedersen, Christian Frederik Larsen, Bertram Madsen, Martin Eeg
Format: Article
Language:English
Published: Nature Portfolio 2023-01-01
Series:Scientific Reports
Online Access:https://doi.org/10.1038/s41598-023-27980-y
_version_ 1797945864355840000
author Mette Pedersen
Christian Frederik Larsen
Bertram Madsen
Martin Eeg
author_facet Mette Pedersen
Christian Frederik Larsen
Bertram Madsen
Martin Eeg
author_sort Mette Pedersen
collection DOAJ
description Abstract The entire glottis has mostly been the focus in the tracking of the vocal folds, both manually and automatically. From a treatment point of view, the various regions of the glottis are of specific interest. The aim of the study was to test if it was possible to supplement an existing convolutional neural network (CNN) with post-network calculations for the localization and quantification of posterior glottal gaps during phonation, usable for vocal fold function analysis of e.g. laryngopharyngeal reflux findings. 30 subjects/videos with insufficient closure in the rear glottal area and 20 normal subjects/videos were selected from our database, recorded with a commercial high-speed video setup (HSV with 4000 frames per second), and segmented with an open-source CNN for validating voice function. We made post-network calculations to localize and quantify the 10% and 50% distance lines from the rear part of the glottis. The results showed a significant difference using the algorithm at the 10% line distance between the two groups of p < 0.0001 and no difference at 50%. These novel results show that it is possible to use post-network calculations on CNNs for the localization and quantification of posterior glottal gaps.
first_indexed 2024-04-10T21:02:53Z
format Article
id doaj.art-2dd3835d540a4a27a9e77d1df4cae741
institution Directory Open Access Journal
issn 2045-2322
language English
last_indexed 2024-04-10T21:02:53Z
publishDate 2023-01-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj.art-2dd3835d540a4a27a9e77d1df4cae7412023-01-22T12:13:08ZengNature PortfolioScientific Reports2045-23222023-01-011311910.1038/s41598-023-27980-yLocalization and quantification of glottal gaps on deep learning segmentation of vocal foldsMette Pedersen0Christian Frederik Larsen1Bertram Madsen2Martin Eeg3Medical CentreCopenhagen Business SchoolME-TAME-TAAbstract The entire glottis has mostly been the focus in the tracking of the vocal folds, both manually and automatically. From a treatment point of view, the various regions of the glottis are of specific interest. The aim of the study was to test if it was possible to supplement an existing convolutional neural network (CNN) with post-network calculations for the localization and quantification of posterior glottal gaps during phonation, usable for vocal fold function analysis of e.g. laryngopharyngeal reflux findings. 30 subjects/videos with insufficient closure in the rear glottal area and 20 normal subjects/videos were selected from our database, recorded with a commercial high-speed video setup (HSV with 4000 frames per second), and segmented with an open-source CNN for validating voice function. We made post-network calculations to localize and quantify the 10% and 50% distance lines from the rear part of the glottis. The results showed a significant difference using the algorithm at the 10% line distance between the two groups of p < 0.0001 and no difference at 50%. These novel results show that it is possible to use post-network calculations on CNNs for the localization and quantification of posterior glottal gaps.https://doi.org/10.1038/s41598-023-27980-y
spellingShingle Mette Pedersen
Christian Frederik Larsen
Bertram Madsen
Martin Eeg
Localization and quantification of glottal gaps on deep learning segmentation of vocal folds
Scientific Reports
title Localization and quantification of glottal gaps on deep learning segmentation of vocal folds
title_full Localization and quantification of glottal gaps on deep learning segmentation of vocal folds
title_fullStr Localization and quantification of glottal gaps on deep learning segmentation of vocal folds
title_full_unstemmed Localization and quantification of glottal gaps on deep learning segmentation of vocal folds
title_short Localization and quantification of glottal gaps on deep learning segmentation of vocal folds
title_sort localization and quantification of glottal gaps on deep learning segmentation of vocal folds
url https://doi.org/10.1038/s41598-023-27980-y
work_keys_str_mv AT mettepedersen localizationandquantificationofglottalgapsondeeplearningsegmentationofvocalfolds
AT christianfrederiklarsen localizationandquantificationofglottalgapsondeeplearningsegmentationofvocalfolds
AT bertrammadsen localizationandquantificationofglottalgapsondeeplearningsegmentationofvocalfolds
AT martineeg localizationandquantificationofglottalgapsondeeplearningsegmentationofvocalfolds