Difference between revisions of "NeoBrainSeg"

From LRDE

(Created page with "== The Challenge of Cerebral MRI in Neonates: A New Method using Mathematical Morphology for the Segmentation of Structures Including DEHSI == ==== Publication ==== The rela...")
 
 
(16 intermediate revisions by 2 users not shown)
Line 1: Line 1:
  +
−
== The Challenge of Cerebral MRI in Neonates: A New Method using Mathematical Morphology for the Segmentation of Structures Including DEHSI ==
  
  +
= The Challenge of Cerebral MRI in Neonates: A New Method using Mathematical Morphology for the Segmentation of Structures Including DEHSI =
   
 
==== Publication ====
  
==== Publication ====
   
−
The related publication is available from [https://www.lrde.epita.fr/wiki/Publications/xu.18.media this page].
 +
The related publication is available from [http://publications.lrde.epita.fr/xu.18.media this page].
   
 
Bibtex entry:
  
Bibtex entry:
−
<nowiki>
 +
<pre>
  +
@Article{xu.18.media,
−  
  +
author = {Yongchao Xu and Baptiste Morel and Sonia Dahdouh and
−
@Article{ xu.18.media,
  
−
author = {Yongchao Xu and Baptiste Morel and Sonia Dahdouh and
 +
\'Elodie Puybareau and Alessio Virz\`i and H\'el\`ene
−
\'Elodie Puybareau and Alessio Virz\`i and H\'el\`ene
 +
Urien and Thierry~G\'eraud and Catherine Adamsbaum and
  +
Isabelle Bloch},
−
Urien and Thierry~G\'eraud and Catherine Adamsbaum and
  
  +
title = {The Challenge of Cerebral Magnetic Resonance Imaging in
−
Isabelle Bloch},
  
  +
Neonates: {A} New Method using Mathematical Morphology
−
title = {The Challenge of Cerebral Magnetic Resonance Imaging in
  
  +
for the Segmentation of Structures Including Diffuse
−
Neonates: {A} New Method using Mathematical Morphology
  
  +
Excessive High Signal Intensities},
−
for the Segmentation of Structures Including Diffuse
  
  +
journal = {Medical Image Analysis},
−
Excessive High Signal Intensities},
  
−
journal = {Medical Image Analysis},
 +
year = 2018,
−
year = 2018,
 +
pages = {1--23},
  +
url = {http://publications.lrde.epita.fr/xu.18.media},
−
pages = {1--23},
  
  +
note = {To appear}
−
abstract = {Preterm birth is a multifactorial condition associated
  
−
with increased morbidity and mortality. Diffuse excessive
  
−
high signal intensity (DEHSI) has been recently described
  
−
on T2-weighted MR sequences in this population and thought
  
−
to be associated with neuropathologies. To date, no robust
  
−
and reproducible method to assess the presence of white
  
−
matter hyperintensities has been developed, perhaps
  
−
explaining the current controversy over their prognostic
  
−
value. The aim of this paper is to propose a new
  
−
semi-automated framework to detect DEHSI on neonatal brain
  
−
MR images having a particular pattern due to the
  
−
physiological lack of complete myelination of the white
  
−
matter. A novel method for semi- automatic segmentation
  
−
of neonatal brain structures and DEHSI, based on
  
−
mathematical morphology and on max-tree representations of
  
−
the images is thus described. It is a mandatory first
  
−
step to identify and clinically assess homogeneous cohorts
  
−
of neonates for DEHSI and/or volume of any other segmented
  
−
structures. Implemented in a user-friendly interface, the
  
−
method makes it straightforward to select relevant markers
  
−
of structures to be segmented, and if needed, apply
  
−
eventually manual corrections. This method responds to the
  
−
increasing need for providing medical experts with
  
−
semi-automatic tools for image analysis, and overcomes the
  
−
limitations of visual analysis alone, prone to
  
−
subjectivity and variability. Experimental results
  
−
demonstrate that the method is accurate, with excellent
  
−
reproducibility and with very few manual corrections
  
−
needed. Although the method was intended initially for
  
−
images acquired at 1.5T, which corresponds to usual
  
−
clinical practice, preliminary results on images acquired
  
−
at 3T suggest that the proposed approach can be
  
−
generalized.},
  
−
url = {http://publications.lrde.epita.fr/xu.18.media.pdf},
  
−
note = {To appear}
  
 
}
  
}
−
</nowiki>
 +
</pre>
   
  +
==== Software ====
  +
  +
[{{SERVER}}/dload/NeoBrainSeg/software_NeoBrainSeg.zip software_NeoBrainSeg.zip] (939Mo)
  +
  +
If this software is used in the context of a scientific publication, please cite the paper mentioned above.
  +
  +
==== Copyright Notice ====
  +
  +
The copyright holders of the main code included in the archive are:
  +
* EPITA Research and Development Laboratory ([http://www.lrde.epita.fr LRDE])
  +
* LTCI Télécom ParisTech
  +
* Faculty of Medicine Bicêtre Hospital APHP
  +
* Faculty of Medicine CHRU Tours.
  +
  +
You are allowed to use these codes and software for research purpose. If so, please specify the following copyright: "Copyright (c) 2018. EPITA Research and Development Laboratory (LRDE)". You are '''not''' allowed to redistribute these codes.
  +
You are '''not''' allowed to redistribute the dehsi.nii example.
  +
  +
  +
<br>
  +
  +
  +
= White Matter Hyperintensities Segmentation In a Few Seconds Using FCN and Transfer Learning =
  +
  +
==== Publication ====
  +
  +
The related publication is available from [http://publications.lrde.epita.fr/xu.18.brainles this page].
  +
  +
Bibtex entry:
  +
<pre>
  +
@InProceedings{xu.18.brainles,
  +
author = {Yongchao Xu and Thierry G{\'e}raud and {\'E}lodie
  +
Puybareau and Isabelle Bloch and Joseph Chazalon},
  +
title = {White Matter Hyperintensities Segmentation In a Few
  +
Seconds Using Fully Convolutional Network and Transfer
  +
Learning},
  +
booktitle = {Brainlesion: Glioma, Multiple Sclerosis, Stroke and
  +
Traumatic Brain Injuries--- 3rd International Workshop,
  +
BrainLes 2017, Held in Conjunction with MICCAI 2017,
  +
Quebec City, QC, Canada, September 14 2017, Revised
  +
Selected Papers},
  +
publisher = {Springer, Cham},
  +
year = {2018},
  +
editor = {A. Crimi and S. Bakas and H. Kuijf and B. Menze and
  +
M. Reyes},
  +
series = {Lecture Notes in Computer Science},
  +
volume = {10670},
  +
pages = {501--514},
  +
doi = {10.1007/978-3-319-75238-9_42},
  +
url = {http://publications.lrde.epita.fr/xu.18.brainles},
  +
}
  +
</pre>
   
 
=== Software ===
  
=== Software ===
  +
[{{SERVER}}/dload/NeoBrainSeg/lrde.tar.zip lrde.tar.zip] (1.3Go)
   
  +
If this software is used in the context of a scientific publication, please cite the paper mentioned above.
−
Available soon...
  
  +
  +
The software is provided as a Docker container. Just run:
  +
<pre>
  +
CONTAINERID=`docker run -dit -v [TEST-ORIG]:/input/orig:ro -v [TEST-PRE]:/input/pre:ro -v /output wmhchallenge/LRDE`
  +
docker exec $CONTAINERID [YOUR-COMMAND]
  +
docker cp $CONTAINERID:/output [RESULT-TEAM]
  +
docker stop $CONTAINERID
  +
docker rm -v $CONTAINERID
  +
</pre>
  +
  +
Two folders are needed in your local folder: the entry data (FLAIR and T1) should be in input/pre, and the result will be written in output. This docker has been done for the MICCAI 2017 challenge: http://wmh.isi.uu.nl/
  +
  +
<br>
  +
  +
= iSeg: MICCAI Grand Challenge on 6-month infant brain MRI Segmentation =
  +
  +
==== Publication ====
  +
  +
A review article of the challenge will be available soon.
  +
  +
==== Software ====
  +
  +
[{{SERVER}}/dload/NeoBrainSeg/Challenge_iSeg.zip Challenge_iSeg.zip] (28.9Mo)
   
 
If this software is used in the context of a scientific publication, please cite the paper mentioned above.
  
If this software is used in the context of a scientific publication, please cite the paper mentioned above.
Line 67: Line 109:
 
==== Copyright Notice ====
  
==== Copyright Notice ====
   
−
[http://www.lrde.epita.fr LRDE] is the copyright holder of all the codes included in the archive.
 +
The copyright holders of the main code included in the archive are:
  +
* EPITA Research and Development Laboratory ([http://www.lrde.epita.fr LRDE])
  +
* LTCI Télécom ParisTech
  +
* Huazhong University of Science and Technology, Wuhan, China
   
 
You are allowed to use these codes and software for research purpose. If so, please specify the following copyright: "Copyright (c) 2018. EPITA Research and Development Laboratory (LRDE)". You are '''not''' allowed to redistribute these codes.
  
You are allowed to use these codes and software for research purpose. If so, please specify the following copyright: "Copyright (c) 2018. EPITA Research and Development Laboratory (LRDE)". You are '''not''' allowed to redistribute these codes.
−
You are '''not''' allowed to redistribute the dehsi.nii example.
  

Latest revision as of 15:49, 27 August 2018

The Challenge of Cerebral MRI in Neonates: A New Method using Mathematical Morphology for the Segmentation of Structures Including DEHSI

Publication

The related publication is available from this page.

Bibtex entry: 

@Article{xu.18.media,
  author = {Yongchao Xu and Baptiste Morel and Sonia Dahdouh and
            \'Elodie Puybareau and Alessio Virz\`i and H\'el\`ene
            Urien and Thierry~G\'eraud and Catherine Adamsbaum and
            Isabelle Bloch},
  title  = {The Challenge of Cerebral Magnetic Resonance Imaging in
            Neonates: {A} New Method using Mathematical Morphology
            for the Segmentation of Structures Including Diffuse
            Excessive High Signal Intensities},
  journal = {Medical Image Analysis},
  year    = 2018,
  pages   = {1--23},
  url     = {http://publications.lrde.epita.fr/xu.18.media},
  note    = {To appear}
}

Software

software_NeoBrainSeg.zip (939Mo)

If this software is used in the context of a scientific publication, please cite the paper mentioned above.

Copyright Notice

The copyright holders of the main code included in the archive are:

  • EPITA Research and Development Laboratory (LRDE)
  • LTCI Télécom ParisTech
  • Faculty of Medicine Bicêtre Hospital APHP
  • Faculty of Medicine CHRU Tours.

You are allowed to use these codes and software for research purpose. If so, please specify the following copyright: "Copyright (c) 2018. EPITA Research and Development Laboratory (LRDE)". You are not allowed to redistribute these codes. You are not allowed to redistribute the dehsi.nii example.




White Matter Hyperintensities Segmentation In a Few Seconds Using FCN and Transfer Learning

Publication

The related publication is available from this page.

Bibtex entry: 

@InProceedings{xu.18.brainles,
  author    = {Yongchao Xu and Thierry G{\'e}raud and {\'E}lodie
               Puybareau and Isabelle Bloch and Joseph Chazalon},
  title     = {White Matter Hyperintensities Segmentation In a Few
               Seconds Using Fully Convolutional Network and Transfer
               Learning},
  booktitle = {Brainlesion: Glioma, Multiple Sclerosis, Stroke and
               Traumatic Brain Injuries--- 3rd International Workshop,
               BrainLes 2017, Held in Conjunction with MICCAI 2017,
               Quebec City, QC, Canada, September 14 2017, Revised
               Selected Papers},
  publisher = {Springer, Cham},
  year      = {2018},
  editor    = {A. Crimi and S. Bakas and H. Kuijf and B. Menze and
               M. Reyes},
  series    = {Lecture Notes in Computer Science},
  volume    = {10670},
  pages     = {501--514},
  doi       = {10.1007/978-3-319-75238-9_42},
  url       = {http://publications.lrde.epita.fr/xu.18.brainles},
}

Software

lrde.tar.zip (1.3Go)

If this software is used in the context of a scientific publication, please cite the paper mentioned above.

The software is provided as a Docker container. Just run: 

CONTAINERID=`docker run -dit -v [TEST-ORIG]:/input/orig:ro -v [TEST-PRE]:/input/pre:ro -v /output wmhchallenge/LRDE`
docker exec $CONTAINERID [YOUR-COMMAND]
docker cp $CONTAINERID:/output [RESULT-TEAM]
docker stop $CONTAINERID
docker rm -v $CONTAINERID

Two folders are needed in your local folder: the entry data (FLAIR and T1) should be in input/pre, and the result will be written in output. This docker has been done for the MICCAI 2017 challenge: http://wmh.isi.uu.nl/


iSeg: MICCAI Grand Challenge on 6-month infant brain MRI Segmentation

Publication

A review article of the challenge will be available soon.

Software

Challenge_iSeg.zip (28.9Mo)

If this software is used in the context of a scientific publication, please cite the paper mentioned above.

Copyright Notice

The copyright holders of the main code included in the archive are:

  • EPITA Research and Development Laboratory (LRDE)
  • LTCI Télécom ParisTech
  • Huazhong University of Science and Technology, Wuhan, China

You are allowed to use these codes and software for research purpose. If so, please specify the following copyright: "Copyright (c) 2018. EPITA Research and Development Laboratory (LRDE)". You are not allowed to redistribute these codes.

Retrieved from "https://www.lrde.epita.fr/index.php?title=NeoBrainSeg&oldid=122328"