Cargando…
Creating a training set for artificial intelligence from initial segmentations of airways
Airways segmentation is important for research about pulmonary disease but require a large amount of time by trained specialists. We used an openly available software to improve airways segmentations obtained from an artificial intelligence (AI) tool and retrained the tool to get a better performanc...
Autores principales: | , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Springer International Publishing
2021
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8627914/ https://www.ncbi.nlm.nih.gov/pubmed/34841480 http://dx.doi.org/10.1186/s41747-021-00247-9 |
_version_ | 1784606912125337600 |
---|---|
author | Dudurych, Ivan Garcia-Uceda, Antonio Saghir, Zaigham Tiddens, Harm A. W. M. Vliegenthart, Rozemarijn de Bruijne, Marleen |
author_facet | Dudurych, Ivan Garcia-Uceda, Antonio Saghir, Zaigham Tiddens, Harm A. W. M. Vliegenthart, Rozemarijn de Bruijne, Marleen |
author_sort | Dudurych, Ivan |
collection | PubMed |
description | Airways segmentation is important for research about pulmonary disease but require a large amount of time by trained specialists. We used an openly available software to improve airways segmentations obtained from an artificial intelligence (AI) tool and retrained the tool to get a better performance. Fifteen initial airway segmentations from low-dose chest computed tomography scans were obtained with a 3D-Unet AI tool previously trained on Danish Lung Cancer Screening Trial and Erasmus-MC Sophia datasets. Segmentations were manually corrected in 3D Slicer. The corrected airway segmentations were used to retrain the 3D-Unet. Airway measurements were automatically obtained and included count, airway length and luminal diameter per generation from the segmentations. Correcting segmentations required 2–4 h per scan. Manually corrected segmentations had more branches (p < 0.001), longer airways (p < 0.001) and smaller luminal diameters (p = 0.004) than initial segmentations. Segmentations from retrained 3D-Unets trended towards more branches and longer airways compared to the initial segmentations. The largest changes were seen in airways from 6th generation onwards. Manual correction results in significantly improved segmentations and is potentially a useful and time-efficient method to improve the AI tool performance on a specific hospital or research dataset. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s41747-021-00247-9. |
format | Online Article Text |
id | pubmed-8627914 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2021 |
publisher | Springer International Publishing |
record_format | MEDLINE/PubMed |
spelling | pubmed-86279142021-12-10 Creating a training set for artificial intelligence from initial segmentations of airways Dudurych, Ivan Garcia-Uceda, Antonio Saghir, Zaigham Tiddens, Harm A. W. M. Vliegenthart, Rozemarijn de Bruijne, Marleen Eur Radiol Exp Technical Note Airways segmentation is important for research about pulmonary disease but require a large amount of time by trained specialists. We used an openly available software to improve airways segmentations obtained from an artificial intelligence (AI) tool and retrained the tool to get a better performance. Fifteen initial airway segmentations from low-dose chest computed tomography scans were obtained with a 3D-Unet AI tool previously trained on Danish Lung Cancer Screening Trial and Erasmus-MC Sophia datasets. Segmentations were manually corrected in 3D Slicer. The corrected airway segmentations were used to retrain the 3D-Unet. Airway measurements were automatically obtained and included count, airway length and luminal diameter per generation from the segmentations. Correcting segmentations required 2–4 h per scan. Manually corrected segmentations had more branches (p < 0.001), longer airways (p < 0.001) and smaller luminal diameters (p = 0.004) than initial segmentations. Segmentations from retrained 3D-Unets trended towards more branches and longer airways compared to the initial segmentations. The largest changes were seen in airways from 6th generation onwards. Manual correction results in significantly improved segmentations and is potentially a useful and time-efficient method to improve the AI tool performance on a specific hospital or research dataset. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s41747-021-00247-9. Springer International Publishing 2021-11-29 /pmc/articles/PMC8627914/ /pubmed/34841480 http://dx.doi.org/10.1186/s41747-021-00247-9 Text en © The Author(s) 2021 https://creativecommons.org/licenses/by/4.0/Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ (https://creativecommons.org/licenses/by/4.0/) . |
spellingShingle | Technical Note Dudurych, Ivan Garcia-Uceda, Antonio Saghir, Zaigham Tiddens, Harm A. W. M. Vliegenthart, Rozemarijn de Bruijne, Marleen Creating a training set for artificial intelligence from initial segmentations of airways |
title | Creating a training set for artificial intelligence from initial segmentations of airways |
title_full | Creating a training set for artificial intelligence from initial segmentations of airways |
title_fullStr | Creating a training set for artificial intelligence from initial segmentations of airways |
title_full_unstemmed | Creating a training set for artificial intelligence from initial segmentations of airways |
title_short | Creating a training set for artificial intelligence from initial segmentations of airways |
title_sort | creating a training set for artificial intelligence from initial segmentations of airways |
topic | Technical Note |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8627914/ https://www.ncbi.nlm.nih.gov/pubmed/34841480 http://dx.doi.org/10.1186/s41747-021-00247-9 |
work_keys_str_mv | AT dudurychivan creatingatrainingsetforartificialintelligencefrominitialsegmentationsofairways AT garciaucedaantonio creatingatrainingsetforartificialintelligencefrominitialsegmentationsofairways AT saghirzaigham creatingatrainingsetforartificialintelligencefrominitialsegmentationsofairways AT tiddensharmawm creatingatrainingsetforartificialintelligencefrominitialsegmentationsofairways AT vliegenthartrozemarijn creatingatrainingsetforartificialintelligencefrominitialsegmentationsofairways AT debruijnemarleen creatingatrainingsetforartificialintelligencefrominitialsegmentationsofairways |