In EDS ansehen

Attention-driven UNet enhancement for accurate segmentation of bacterial spore outgrowth in microscopy images.

Gespeichert in:
Bibliographische Detailangaben
Titel: Attention-driven UNet enhancement for accurate segmentation of bacterial spore outgrowth in microscopy images.
Autoren: Qamar S; Department of Physics, Umeå University, 90187, Umeå, Sweden.; Integrated Science Lab, Umeå University, 90187, Umeå, Sweden.; Faculty of Computing and IT, Sohar University, 311, Sohar, Oman., Malyshev D; Department of Physics, Umeå University, 90187, Umeå, Sweden., Öberg R; Department of Physics, Umeå University, 90187, Umeå, Sweden., Nilsson DPG; Department of Physics, Umeå University, 90187, Umeå, Sweden., Andersson M; Department of Physics, Umeå University, 90187, Umeå, Sweden. magnus.andersson@umu.se.; Integrated Science Lab, Umeå University, 90187, Umeå, Sweden. magnus.andersson@umu.se.; Umeå Centre for Microbial Research (UCMR), Umeå University, 90187, Umeå, Sweden. magnus.andersson@umu.se.
Quelle: Scientific reports [Sci Rep] 2025 Jun 20; Vol. 15 (1), pp. 20177. Date of Electronic Publication: 2025 Jun 20.
Publikationsart: Journal Article
Sprache: English
Info zur Zeitschrift: Publisher: Nature Publishing Group Country of Publication: England NLM ID: 101563288 Publication Model: Electronic Cited Medium: Internet ISSN: 2045-2322 (Electronic) Linking ISSN: 20452322 NLM ISO Abbreviation: Sci Rep Subsets: MEDLINE
Imprint Name(s): Original Publication: London : Nature Publishing Group, copyright 2011-
MeSH-Schlagworte: Spores, Bacterial*/growth & development , Image Processing, Computer-Assisted*/methods , Microscopy*/methods, Deep Learning ; Algorithms
Abstract: Analyzing microscopy images of large growing cell samples using traditional methods is a complex and time-consuming process. In this work, we have developed an attention-driven UNet-enhanced model using deep learning techniques to efficiently quantify the position, area, and circularity of bacterial spores and vegetative cells from images containing more than 10,000 bacterial cells. Our attention-driven UNet algorithm has an accuracy of 96%, precision of 82%, sensitivity of 81%, and specificity of 98%. Therefore, it can segment cells at a level comparable to manual annotation. We demonstrate the efficacy of this model by applying it to a live-dead decontamination assay. The model is provided in three formats: Python code, a Binder that operates within a web browser without needing installation, and a Flask Web application for local use.
(© 2025. The Author(s).)
Competing Interests: Declarations. Ethical approval: This study did not involve animals or human participants. Competing interests: The authors declare that they have no conflicts of interest. We declare that the authors have no competing interests as defined by Springer or other interests that could be perceived to influence the results and/or discussion reported in this paper.
References: Sci Rep. 2023 Oct 31;13(1):18758. (PMID: 37907463)
Bioinformatics. 2009 Jun 1;25(11):1463-5. (PMID: 19346324)
Food Microbiol. 2010 May;27(3):347-55. (PMID: 20227599)
Nat Methods. 2012 Jun 28;9(7):676-82. (PMID: 22743772)
Sci Data. 2023 Oct 4;10(1):677. (PMID: 37794110)
Med Image Anal. 2019 Apr;53:197-207. (PMID: 30802813)
Nat Methods. 2021 Jan;18(1):43-45. (PMID: 33398191)
Mol Biol Cell. 2023 May 15;34(6):ar50. (PMID: 36515991)
Biofouling. 2019 Feb;35(2):204-216. (PMID: 30950292)
Anal Chem. 2021 Feb 16;93(6):3146-3153. (PMID: 33523636)
Front Comput Neurosci. 2020 Apr 16;14:29. (PMID: 32372937)
Appl Environ Microbiol. 2016 Jan 22;82(7):2031-2038. (PMID: 26801572)
Elife. 2021 Sep 09;10:. (PMID: 34498586)
PLoS One. 2013;8(3):e58972. (PMID: 23536843)
Opt Lett. 2018 May 1;43(9):1990-1993. (PMID: 29714728)
Sci Rep. 2023 Mar 1;13(1):3483. (PMID: 36859457)
Nat Methods. 2019 Dec;16(12):1226-1232. (PMID: 31570887)
Anal Chem. 2009 May 15;81(10):4035-42. (PMID: 19374431)
Nat Methods. 2022 Nov;19(11):1438-1448. (PMID: 36253643)
Appl Environ Microbiol. 2004 Sep;70(9):5343-8. (PMID: 15345419)
Nat Methods. 2021 Jan;18(1):100-106. (PMID: 33318659)
BMC Microbiol. 2023 Mar 6;23(1):59. (PMID: 36879193)
Biophys J. 2006 Oct 1;91(7):2717-25. (PMID: 16844748)
IEEE Trans Med Imaging. 2022 Apr;41(4):965-976. (PMID: 34813472)
J Big Data. 2021;8(1):101. (PMID: 34306963)
FEMS Microbiol Rev. 2021 Aug 17;45(4):. (PMID: 33242074)
Infect Control Hosp Epidemiol. 2007 Aug;28(8):920-5. (PMID: 17620238)
J Hosp Infect. 2022 Apr;122:140-147. (PMID: 35077809)
Mol Microbiol. 2020 Jul;114(1):140-150. (PMID: 32190923)
J Biophotonics. 2022 Aug;15(8):e202200081. (PMID: 35538633)
Science. 2003 Apr 4;300(5616):82-6. (PMID: 12677057)
Nat Protoc. 2011 Dec 15;7(1):80-8. (PMID: 22179594)
Grant Information: JCK-2129.3 Kempestiftelserna; 2022-06725 NAISS
Contributed Indexing: Keywords: Contamination; Deep learning; Spores
Entry Date(s): Date Created: 20250620 Date Completed: 20250624 Latest Revision: 20250624
Update Code: 20250625
PubMed Central ID: PMC12181384
DOI: 10.1038/s41598-025-05900-6
PMID: 40542045
Datenbank: MEDLINE
Beschreibung
Abstract:Analyzing microscopy images of large growing cell samples using traditional methods is a complex and time-consuming process. In this work, we have developed an attention-driven UNet-enhanced model using deep learning techniques to efficiently quantify the position, area, and circularity of bacterial spores and vegetative cells from images containing more than 10,000 bacterial cells. Our attention-driven UNet algorithm has an accuracy of 96%, precision of 82%, sensitivity of 81%, and specificity of 98%. Therefore, it can segment cells at a level comparable to manual annotation. We demonstrate the efficacy of this model by applying it to a live-dead decontamination assay. The model is provided in three formats: Python code, a Binder that operates within a web browser without needing installation, and a Flask Web application for local use.<br /> (© 2025. The Author(s).)
ISSN:2045-2322
DOI:10.1038/s41598-025-05900-6