Automating cell counting in fluorescent microscopy through deep learning with c-ResUnet

Abstract Counting cells in fluorescent microscopy is a tedious, time-consuming task that researchers have to accomplish to assess the effects of different experimental conditions on biological structures of interest. Although such objects are generally easy to identify, the process of manually annot...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Roberto Morelli, Luca Clissa, Roberto Amici, Matteo Cerri, Timna Hitrec, Marco Luppi, Lorenzo Rinaldi, Fabio Squarcio, Antonio Zoccoli
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/6ead4c94f6484e7f9793ae43d1f5239f
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:6ead4c94f6484e7f9793ae43d1f5239f
record_format	dspace
spelling	oai:doaj.org-article:6ead4c94f6484e7f9793ae43d1f5239f2021-11-28T12:20:21ZAutomating cell counting in fluorescent microscopy through deep learning with c-ResUnet10.1038/s41598-021-01929-52045-2322https://doaj.org/article/6ead4c94f6484e7f9793ae43d1f5239f2021-11-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-01929-5https://doaj.org/toc/2045-2322Abstract Counting cells in fluorescent microscopy is a tedious, time-consuming task that researchers have to accomplish to assess the effects of different experimental conditions on biological structures of interest. Although such objects are generally easy to identify, the process of manually annotating cells is sometimes subject to fatigue errors and suffers from arbitrariness due to the operator’s interpretation of the borderline cases. We propose a Deep Learning approach that exploits a fully-convolutional network in a binary segmentation fashion to localize the objects of interest. Counts are then retrieved as the number of detected items. Specifically, we introduce a Unet-like architecture, cell ResUnet (c-ResUnet), and compare its performance against 3 similar architectures. In addition, we evaluate through ablation studies the impact of two design choices, (i) artifacts oversampling and (ii) weight maps that penalize the errors on cells boundaries increasingly with overcrowding. In summary, the c-ResUnet outperforms the competitors with respect to both detection and counting metrics (respectively, $$F_1$$ F 1 score = 0.81 and MAE = 3.09). Also, the introduction of weight maps contribute to enhance performances, especially in presence of clumping cells, artifacts and confounding biological structures. Posterior qualitative assessment by domain experts corroborates previous results, suggesting human-level performance inasmuch even erroneous predictions seem to fall within the limits of operator interpretation. Finally, we release the pre-trained model and the annotated dataset to foster research in this and related fields.Roberto MorelliLuca ClissaRoberto AmiciMatteo CerriTimna HitrecMarco LuppiLorenzo RinaldiFabio SquarcioAntonio ZoccoliNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-11 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Roberto Morelli Luca Clissa Roberto Amici Matteo Cerri Timna Hitrec Marco Luppi Lorenzo Rinaldi Fabio Squarcio Antonio Zoccoli Automating cell counting in fluorescent microscopy through deep learning with c-ResUnet
description	Abstract Counting cells in fluorescent microscopy is a tedious, time-consuming task that researchers have to accomplish to assess the effects of different experimental conditions on biological structures of interest. Although such objects are generally easy to identify, the process of manually annotating cells is sometimes subject to fatigue errors and suffers from arbitrariness due to the operator’s interpretation of the borderline cases. We propose a Deep Learning approach that exploits a fully-convolutional network in a binary segmentation fashion to localize the objects of interest. Counts are then retrieved as the number of detected items. Specifically, we introduce a Unet-like architecture, cell ResUnet (c-ResUnet), and compare its performance against 3 similar architectures. In addition, we evaluate through ablation studies the impact of two design choices, (i) artifacts oversampling and (ii) weight maps that penalize the errors on cells boundaries increasingly with overcrowding. In summary, the c-ResUnet outperforms the competitors with respect to both detection and counting metrics (respectively, $$F_1$$ F 1 score = 0.81 and MAE = 3.09). Also, the introduction of weight maps contribute to enhance performances, especially in presence of clumping cells, artifacts and confounding biological structures. Posterior qualitative assessment by domain experts corroborates previous results, suggesting human-level performance inasmuch even erroneous predictions seem to fall within the limits of operator interpretation. Finally, we release the pre-trained model and the annotated dataset to foster research in this and related fields.
format	article
author	Roberto Morelli Luca Clissa Roberto Amici Matteo Cerri Timna Hitrec Marco Luppi Lorenzo Rinaldi Fabio Squarcio Antonio Zoccoli
author_facet	Roberto Morelli Luca Clissa Roberto Amici Matteo Cerri Timna Hitrec Marco Luppi Lorenzo Rinaldi Fabio Squarcio Antonio Zoccoli
author_sort	Roberto Morelli
title	Automating cell counting in fluorescent microscopy through deep learning with c-ResUnet
title_short	Automating cell counting in fluorescent microscopy through deep learning with c-ResUnet
title_full	Automating cell counting in fluorescent microscopy through deep learning with c-ResUnet
title_fullStr	Automating cell counting in fluorescent microscopy through deep learning with c-ResUnet
title_full_unstemmed	Automating cell counting in fluorescent microscopy through deep learning with c-ResUnet
title_sort	automating cell counting in fluorescent microscopy through deep learning with c-resunet
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/6ead4c94f6484e7f9793ae43d1f5239f
work_keys_str_mv	AT robertomorelli automatingcellcountinginfluorescentmicroscopythroughdeeplearningwithcresunet AT lucaclissa automatingcellcountinginfluorescentmicroscopythroughdeeplearningwithcresunet AT robertoamici automatingcellcountinginfluorescentmicroscopythroughdeeplearningwithcresunet AT matteocerri automatingcellcountinginfluorescentmicroscopythroughdeeplearningwithcresunet AT timnahitrec automatingcellcountinginfluorescentmicroscopythroughdeeplearningwithcresunet AT marcoluppi automatingcellcountinginfluorescentmicroscopythroughdeeplearningwithcresunet AT lorenzorinaldi automatingcellcountinginfluorescentmicroscopythroughdeeplearningwithcresunet AT fabiosquarcio automatingcellcountinginfluorescentmicroscopythroughdeeplearningwithcresunet AT antoniozoccoli automatingcellcountinginfluorescentmicroscopythroughdeeplearningwithcresunet
_version_	1718408022351413248

Automating cell counting in fluorescent microscopy through deep learning with c-ResUnet

Ejemplares similares