Difference between revisions of "Publications/puybareau.19.cbm"

From LRDE

(Created page with "{{Publication | published = true | date = 2019-01-22 | authors = Diane Genest, Élodie Puybareau, Marc Léonard, Jean Cousty, Noémie De Crozé, Hugues Talbot | title = High t...")
 
 
(One intermediate revision by the same user not shown)
Line 3: Line 3:
 
| date = 2019-01-22
 
| date = 2019-01-22
 
| authors = Diane Genest, Élodie Puybareau, Marc Léonard, Jean Cousty, Noémie De Crozé, Hugues Talbot
 
| authors = Diane Genest, Élodie Puybareau, Marc Léonard, Jean Cousty, Noémie De Crozé, Hugues Talbot
| title = High throughput automated detection of axial malformations in Medaka embryo
+
| title = High Throughput Automated Detection of Axial Malformations in Medaka Embryo
 
| journal = Computers in Biology and Medicine
 
| journal = Computers in Biology and Medicine
 
| pages = 157 to 168
 
| pages = 157 to 168
  +
| volume = 105
 
| lrdeprojects = Olena
 
| lrdeprojects = Olena
 
| abstract = Fish embryo models are widely used as screening tools to assess the efficacy and/or toxicity of chemicals. This assessment involves the analysis of embryo morphological abnormalities. In this article, we propose a multi-scale pipeline to allow automated classification of fish embryos (Medaka: Oryzias latipes) based on the presence or absence of spine malformations. The proposed pipeline relies on the acquisition of fish embryo 2D images, on feature extraction based on mathematical morphology operators and on machine learning classification. After image acquisitionsegmentation tools are used to detect the embryo before analysing several morphological features. An approach based on machine learning is then applied to these features to automatically classify embryos according to the presence of axial malformations. We built and validated our learning model on 1459 images with a 10-fold cross-validation by comparison with the gold standard of 3D observations performed under a microscope by a trained operator. Our pipeline results in correct classification in 85% of the cases included in the database. This percentage is similar to the percentage of success of a trained human operator working on 2D images. The key benefit of our approach is the low computational cost of our image analysis pipelinewhich guarantees optimal throughput analysis..
 
| abstract = Fish embryo models are widely used as screening tools to assess the efficacy and/or toxicity of chemicals. This assessment involves the analysis of embryo morphological abnormalities. In this article, we propose a multi-scale pipeline to allow automated classification of fish embryos (Medaka: Oryzias latipes) based on the presence or absence of spine malformations. The proposed pipeline relies on the acquisition of fish embryo 2D images, on feature extraction based on mathematical morphology operators and on machine learning classification. After image acquisitionsegmentation tools are used to detect the embryo before analysing several morphological features. An approach based on machine learning is then applied to these features to automatically classify embryos according to the presence of axial malformations. We built and validated our learning model on 1459 images with a 10-fold cross-validation by comparison with the gold standard of 3D observations performed under a microscope by a trained operator. Our pipeline results in correct classification in 85% of the cases included in the database. This percentage is similar to the percentage of success of a trained human operator working on 2D images. The key benefit of our approach is the low computational cost of our image analysis pipelinewhich guarantees optimal throughput analysis..
Line 12: Line 13:
 
| type = article
 
| type = article
 
| id = puybareau.19.cbm
 
| id = puybareau.19.cbm
  +
| identifier = doi:10.1016/j.compbiomed.2018.12.016
 
| bibtex =
 
| bibtex =
 
@Article<nowiki>{</nowiki> puybareau.19.cbm,
 
@Article<nowiki>{</nowiki> puybareau.19.cbm,
 
author = <nowiki>{</nowiki>Diane Genest and \'Elodie Puybareau and Marc L\'eonard and
 
author = <nowiki>{</nowiki>Diane Genest and \'Elodie Puybareau and Marc L\'eonard and
 
Jean Cousty and No\'emie De Croz\'e and Hugues Talbot<nowiki>}</nowiki>,
 
Jean Cousty and No\'emie De Croz\'e and Hugues Talbot<nowiki>}</nowiki>,
title = <nowiki>{</nowiki>High throughput automated detection of axial malformations
+
title = <nowiki>{</nowiki>High Throughput Automated Detection of Axial Malformations
in Medaka embryo<nowiki>}</nowiki>,
+
in <nowiki>{</nowiki>M<nowiki>}</nowiki>edaka Embryo<nowiki>}</nowiki>,
 
journal = <nowiki>{</nowiki>Computers in Biology and Medicine<nowiki>}</nowiki>,
 
journal = <nowiki>{</nowiki>Computers in Biology and Medicine<nowiki>}</nowiki>,
 
year = 2019,
 
year = 2019,
  +
month = feb,
 
pages = <nowiki>{</nowiki>157--168<nowiki>}</nowiki>,
 
pages = <nowiki>{</nowiki>157--168<nowiki>}</nowiki>,
  +
volume = <nowiki>{</nowiki>105<nowiki>}</nowiki>,
 
abstract = <nowiki>{</nowiki>Fish embryo models are widely used as screening tools to
 
abstract = <nowiki>{</nowiki>Fish embryo models are widely used as screening tools to
 
assess the efficacy and/or toxicity of chemicals. This
 
assess the efficacy and/or toxicity of chemicals. This
Line 44: Line 48:
 
working on 2D images. The key benefit of our approach is
 
working on 2D images. The key benefit of our approach is
 
the low computational cost of our image analysis pipeline,
 
the low computational cost of our image analysis pipeline,
which guarantees optimal throughput analysis..<nowiki>}</nowiki>
+
which guarantees optimal throughput analysis..<nowiki>}</nowiki>,
  +
doi = <nowiki>{</nowiki>10.1016/j.compbiomed.2018.12.016<nowiki>}</nowiki>
 
<nowiki>}</nowiki>
 
<nowiki>}</nowiki>
   

Latest revision as of 16:01, 27 May 2021

Abstract

Fish embryo models are widely used as screening tools to assess the efficacy and/or toxicity of chemicals. This assessment involves the analysis of embryo morphological abnormalities. In this article, we propose a multi-scale pipeline to allow automated classification of fish embryos (Medaka: Oryzias latipes) based on the presence or absence of spine malformations. The proposed pipeline relies on the acquisition of fish embryo 2D images, on feature extraction based on mathematical morphology operators and on machine learning classification. After image acquisitionsegmentation tools are used to detect the embryo before analysing several morphological features. An approach based on machine learning is then applied to these features to automatically classify embryos according to the presence of axial malformations. We built and validated our learning model on 1459 images with a 10-fold cross-validation by comparison with the gold standard of 3D observations performed under a microscope by a trained operator. Our pipeline results in correct classification in 85% of the cases included in the database. This percentage is similar to the percentage of success of a trained human operator working on 2D images. The key benefit of our approach is the low computational cost of our image analysis pipelinewhich guarantees optimal throughput analysis..


Bibtex (lrde.bib)

@Article{	  puybareau.19.cbm,
  author	= {Diane Genest and \'Elodie Puybareau and Marc L\'eonard and
		  Jean Cousty and No\'emie De Croz\'e and Hugues Talbot},
  title		= {High Throughput Automated Detection of Axial Malformations
		  in {M}edaka Embryo},
  journal	= {Computers in Biology and Medicine},
  year		= 2019,
  month		= feb,
  pages		= {157--168},
  volume	= {105},
  abstract	= {Fish embryo models are widely used as screening tools to
		  assess the efficacy and/or toxicity of chemicals. This
		  assessment involves the analysis of embryo morphological
		  abnormalities. In this article, we propose a multi-scale
		  pipeline to allow automated classification of fish embryos
		  (Medaka: Oryzias latipes) based on the presence or absence
		  of spine malformations. The proposed pipeline relies on the
		  acquisition of fish embryo 2D images, on feature extraction
		  based on mathematical morphology operators and on machine
		  learning classification. After image acquisition,
		  segmentation tools are used to detect the embryo before
		  analysing several morphological features. An approach based
		  on machine learning is then applied to these features to
		  automatically classify embryos according to the presence of
		  axial malformations. We built and validated our learning
		  model on 1459 images with a 10-fold cross-validation by
		  comparison with the gold standard of 3D observations
		  performed under a microscope by a trained operator. Our
		  pipeline results in correct classification in 85\% of the
		  cases included in the database. This percentage is similar
		  to the percentage of success of a trained human operator
		  working on 2D images. The key benefit of our approach is
		  the low computational cost of our image analysis pipeline,
		  which guarantees optimal throughput analysis..},
  doi		= {10.1016/j.compbiomed.2018.12.016}
}