Cargando…

Classifiers for Ischemic Stroke Lesion Segmentation: A Comparison Study

MOTIVATION: Ischemic stroke, triggered by an obstruction in the cerebral blood supply, leads to infarction of the affected brain tissue. An accurate and reproducible automatic segmentation is of high interest, since the lesion volume is an important end-point for clinical trials. However, various fa...

Descripción completa

Detalles Bibliográficos
Autores principales: Maier, Oskar, Schröder, Christoph, Forkert, Nils Daniel, Martinetz, Thomas, Handels, Heinz
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Public Library of Science 2015
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4687679/
https://www.ncbi.nlm.nih.gov/pubmed/26672989
http://dx.doi.org/10.1371/journal.pone.0145118
_version_ 1782406656004456448
author Maier, Oskar
Schröder, Christoph
Forkert, Nils Daniel
Martinetz, Thomas
Handels, Heinz
author_facet Maier, Oskar
Schröder, Christoph
Forkert, Nils Daniel
Martinetz, Thomas
Handels, Heinz
author_sort Maier, Oskar
collection PubMed
description MOTIVATION: Ischemic stroke, triggered by an obstruction in the cerebral blood supply, leads to infarction of the affected brain tissue. An accurate and reproducible automatic segmentation is of high interest, since the lesion volume is an important end-point for clinical trials. However, various factors, such as the high variance in lesion shape, location and appearance, render it a difficult task. METHODS: In this article, nine classification methods (e.g. Generalized Linear Models, Random Decision Forests and Convolutional Neural Networks) are evaluated and compared with each other using 37 multiparametric MRI datasets of ischemic stroke patients in the sub-acute phase in terms of their accuracy and reliability for ischemic stroke lesion segmentation. Within this context, a multi-spectral classification approach is compared against mono-spectral classification performance using only FLAIR MRI datasets and two sets of expert segmentations are used for inter-observer agreement evaluation. RESULTS AND CONCLUSION: The results of this study reveal that high-level machine learning methods lead to significantly better segmentation results compared to the rather simple classification methods, pointing towards a difficult non-linear problem. The overall best segmentation results were achieved by a Random Decision Forest and a Convolutional Neural Networks classification approach, even outperforming all previously published results. However, none of the methods tested in this work are capable of achieving results in the range of the human observer agreement and the automatic ischemic stroke lesion segmentation remains a complicated problem that needs to be explored in more detail to improve the segmentation results.
format Online
Article
Text
id pubmed-4687679
institution National Center for Biotechnology Information
language English
publishDate 2015
publisher Public Library of Science
record_format MEDLINE/PubMed
spelling pubmed-46876792015-12-31 Classifiers for Ischemic Stroke Lesion Segmentation: A Comparison Study Maier, Oskar Schröder, Christoph Forkert, Nils Daniel Martinetz, Thomas Handels, Heinz PLoS One Research Article MOTIVATION: Ischemic stroke, triggered by an obstruction in the cerebral blood supply, leads to infarction of the affected brain tissue. An accurate and reproducible automatic segmentation is of high interest, since the lesion volume is an important end-point for clinical trials. However, various factors, such as the high variance in lesion shape, location and appearance, render it a difficult task. METHODS: In this article, nine classification methods (e.g. Generalized Linear Models, Random Decision Forests and Convolutional Neural Networks) are evaluated and compared with each other using 37 multiparametric MRI datasets of ischemic stroke patients in the sub-acute phase in terms of their accuracy and reliability for ischemic stroke lesion segmentation. Within this context, a multi-spectral classification approach is compared against mono-spectral classification performance using only FLAIR MRI datasets and two sets of expert segmentations are used for inter-observer agreement evaluation. RESULTS AND CONCLUSION: The results of this study reveal that high-level machine learning methods lead to significantly better segmentation results compared to the rather simple classification methods, pointing towards a difficult non-linear problem. The overall best segmentation results were achieved by a Random Decision Forest and a Convolutional Neural Networks classification approach, even outperforming all previously published results. However, none of the methods tested in this work are capable of achieving results in the range of the human observer agreement and the automatic ischemic stroke lesion segmentation remains a complicated problem that needs to be explored in more detail to improve the segmentation results. Public Library of Science 2015-12-16 /pmc/articles/PMC4687679/ /pubmed/26672989 http://dx.doi.org/10.1371/journal.pone.0145118 Text en © 2015 Maier et al http://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
spellingShingle Research Article
Maier, Oskar
Schröder, Christoph
Forkert, Nils Daniel
Martinetz, Thomas
Handels, Heinz
Classifiers for Ischemic Stroke Lesion Segmentation: A Comparison Study
title Classifiers for Ischemic Stroke Lesion Segmentation: A Comparison Study
title_full Classifiers for Ischemic Stroke Lesion Segmentation: A Comparison Study
title_fullStr Classifiers for Ischemic Stroke Lesion Segmentation: A Comparison Study
title_full_unstemmed Classifiers for Ischemic Stroke Lesion Segmentation: A Comparison Study
title_short Classifiers for Ischemic Stroke Lesion Segmentation: A Comparison Study
title_sort classifiers for ischemic stroke lesion segmentation: a comparison study
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4687679/
https://www.ncbi.nlm.nih.gov/pubmed/26672989
http://dx.doi.org/10.1371/journal.pone.0145118
work_keys_str_mv AT maieroskar classifiersforischemicstrokelesionsegmentationacomparisonstudy
AT schroderchristoph classifiersforischemicstrokelesionsegmentationacomparisonstudy
AT forkertnilsdaniel classifiersforischemicstrokelesionsegmentationacomparisonstudy
AT martinetzthomas classifiersforischemicstrokelesionsegmentationacomparisonstudy
AT handelsheinz classifiersforischemicstrokelesionsegmentationacomparisonstudy