Image Analysis Pipeline for Automatic Karyotyping
Authors: Izaro Goienetxea and Iñigo Barandiaran and Carlos Jauquicoa and Grégory Maclair and Manuel Graña
Date: 28.03.2012
Abstract
The karyotyping step is essential in the genetic diagnosis process, since it allows the genetician to see and interpret patient’s chromosomes. Today, this step of karyotyping is a time-cost procedure, especially the part that consists in segmenting and classifying the chromosomes by pairs. This paper presents an image analysis pipeline of banded human chromosomes for automated karyotyping. The proposed pipeline is composed of three different stages: an image segmentation step, a feature extraction procedure and a final pattern classification task. Two different approaches for the final classification stage were studied, and different classifiers were compared. The obtained results shows that Random Forest classifier combined with a two step classification approach can be considered as an efficient and accurate method for karyotyping.
BIB_text
@Article {
author = {Izaro Goienetxea and Iñigo Barandiaran and Carlos Jauquicoa and Grégory Maclair and Manuel Graña},
title = {Image Analysis Pipeline for Automatic Karyotyping},
pages = {392-403},
volume = {7209},
keywds = {
Image Analysis, Kariotyping, Classifier Ensembles
}
abstract = {
The karyotyping step is essential in the genetic diagnosis process, since it allows the genetician to see and interpret patient’s chromosomes. Today, this step of karyotyping is a time-cost procedure, especially the part that consists in segmenting and classifying the chromosomes by pairs. This paper presents an image analysis pipeline of banded human chromosomes for automated karyotyping. The proposed pipeline is composed of three different stages: an image segmentation step, a feature extraction procedure and a final pattern classification task. Two different approaches for the final classification stage were studied, and different classifiers were compared. The obtained results shows that Random Forest classifier combined with a two step classification approach can be considered as an efficient and accurate method for karyotyping.
}
isbn = {978-3-642-28930-9},
isi = {1},
date = {2012-03-28},
year = {2012},
}
author = {Izaro Goienetxea and Iñigo Barandiaran and Carlos Jauquicoa and Grégory Maclair and Manuel Graña},
title = {Image Analysis Pipeline for Automatic Karyotyping},
pages = {392-403},
volume = {7209},
keywds = {
Image Analysis, Kariotyping, Classifier Ensembles
}
abstract = {
The karyotyping step is essential in the genetic diagnosis process, since it allows the genetician to see and interpret patient’s chromosomes. Today, this step of karyotyping is a time-cost procedure, especially the part that consists in segmenting and classifying the chromosomes by pairs. This paper presents an image analysis pipeline of banded human chromosomes for automated karyotyping. The proposed pipeline is composed of three different stages: an image segmentation step, a feature extraction procedure and a final pattern classification task. Two different approaches for the final classification stage were studied, and different classifiers were compared. The obtained results shows that Random Forest classifier combined with a two step classification approach can be considered as an efficient and accurate method for karyotyping.
}
isbn = {978-3-642-28930-9},
isi = {1},
date = {2012-03-28},
year = {2012},
}
