Automating the assessment of biofouling in images using expert agreement as a gold standard

Abstract Biofouling is the accumulation of organisms on surfaces immersed in water. It is of particular concern to the international shipping industry because it increases fuel costs and presents a biosecurity risk by providing a pathway for non-indigenous marine species to establish in new areas. T...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Nathaniel J. Bloomfield, Susan Wei, Bartholomew A. Woodham, Peter Wilkinson, Andrew P. Robinson
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/425ad8b1fca845df947b8a53c6362581
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:425ad8b1fca845df947b8a53c6362581
record_format	dspace
spelling	oai:doaj.org-article:425ad8b1fca845df947b8a53c63625812021-12-02T10:44:22ZAutomating the assessment of biofouling in images using expert agreement as a gold standard10.1038/s41598-021-81011-22045-2322https://doaj.org/article/425ad8b1fca845df947b8a53c63625812021-02-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-81011-2https://doaj.org/toc/2045-2322Abstract Biofouling is the accumulation of organisms on surfaces immersed in water. It is of particular concern to the international shipping industry because it increases fuel costs and presents a biosecurity risk by providing a pathway for non-indigenous marine species to establish in new areas. There is growing interest within jurisdictions to strengthen biofouling risk-management regulations, but it is expensive to conduct in-water inspections and assess the collected data to determine the biofouling state of vessel hulls. Machine learning is well suited to tackle the latter challenge, and here we apply deep learning to automate the classification of images from in-water inspections to identify the presence and severity of fouling. We combined several datasets to obtain over 10,000 images collected from in-water surveys which were annotated by a group biofouling experts. We compared the annotations from three experts on a 120-sample subset of these images, and found that they showed 89% agreement (95% CI: 87–92%). Subsequent labelling of the whole dataset by one of these experts achieved similar levels of agreement with this group of experts, which we defined as performing at most 5% worse (p $$=$$ = 0.009–0.054). Using these expert labels, we were able to train a deep learning model that also agreed similarly with the group of experts (p $$=$$ = 0.001–0.014), demonstrating that automated analysis of biofouling in images is feasible and effective using this method.Nathaniel J. BloomfieldSusan WeiBartholomew A. WoodhamPeter WilkinsonAndrew P. RobinsonNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-10 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Nathaniel J. Bloomfield Susan Wei Bartholomew A. Woodham Peter Wilkinson Andrew P. Robinson Automating the assessment of biofouling in images using expert agreement as a gold standard
description	Abstract Biofouling is the accumulation of organisms on surfaces immersed in water. It is of particular concern to the international shipping industry because it increases fuel costs and presents a biosecurity risk by providing a pathway for non-indigenous marine species to establish in new areas. There is growing interest within jurisdictions to strengthen biofouling risk-management regulations, but it is expensive to conduct in-water inspections and assess the collected data to determine the biofouling state of vessel hulls. Machine learning is well suited to tackle the latter challenge, and here we apply deep learning to automate the classification of images from in-water inspections to identify the presence and severity of fouling. We combined several datasets to obtain over 10,000 images collected from in-water surveys which were annotated by a group biofouling experts. We compared the annotations from three experts on a 120-sample subset of these images, and found that they showed 89% agreement (95% CI: 87–92%). Subsequent labelling of the whole dataset by one of these experts achieved similar levels of agreement with this group of experts, which we defined as performing at most 5% worse (p $$=$$ = 0.009–0.054). Using these expert labels, we were able to train a deep learning model that also agreed similarly with the group of experts (p $$=$$ = 0.001–0.014), demonstrating that automated analysis of biofouling in images is feasible and effective using this method.
format	article
author	Nathaniel J. Bloomfield Susan Wei Bartholomew A. Woodham Peter Wilkinson Andrew P. Robinson
author_facet	Nathaniel J. Bloomfield Susan Wei Bartholomew A. Woodham Peter Wilkinson Andrew P. Robinson
author_sort	Nathaniel J. Bloomfield
title	Automating the assessment of biofouling in images using expert agreement as a gold standard
title_short	Automating the assessment of biofouling in images using expert agreement as a gold standard
title_full	Automating the assessment of biofouling in images using expert agreement as a gold standard
title_fullStr	Automating the assessment of biofouling in images using expert agreement as a gold standard
title_full_unstemmed	Automating the assessment of biofouling in images using expert agreement as a gold standard
title_sort	automating the assessment of biofouling in images using expert agreement as a gold standard
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/425ad8b1fca845df947b8a53c6362581
work_keys_str_mv	AT nathanieljbloomfield automatingtheassessmentofbiofoulinginimagesusingexpertagreementasagoldstandard AT susanwei automatingtheassessmentofbiofoulinginimagesusingexpertagreementasagoldstandard AT bartholomewawoodham automatingtheassessmentofbiofoulinginimagesusingexpertagreementasagoldstandard AT peterwilkinson automatingtheassessmentofbiofoulinginimagesusingexpertagreementasagoldstandard AT andrewprobinson automatingtheassessmentofbiofoulinginimagesusingexpertagreementasagoldstandard
_version_	1718396758520758272

Automating the assessment of biofouling in images using expert agreement as a gold standard

Ejemplares similares