Deep Learning for Fully-Automated Localization and Segmentation of Rectal Cancer on Multiparametric MR

Abstract Multiparametric Magnetic Resonance Imaging (MRI) can provide detailed information of the physical characteristics of rectum tumours. Several investigations suggest that volumetric analyses on anatomical and functional MRI contain clinically valuable information. However, manual delineation...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Stefano Trebeschi, Joost J. M. van Griethuysen, Doenja M. J. Lambregts, Max J. Lahaye, Chintan Parmar, Frans C. H. Bakers, Nicky H. G. M. Peters, Regina G. H. Beets-Tan, Hugo J. W. L. Aerts
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2017
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/530f2a985a074fb7bdfb89c6da9063f9
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:530f2a985a074fb7bdfb89c6da9063f9
record_format	dspace
spelling	oai:doaj.org-article:530f2a985a074fb7bdfb89c6da9063f92021-12-02T12:30:53ZDeep Learning for Fully-Automated Localization and Segmentation of Rectal Cancer on Multiparametric MR10.1038/s41598-017-05728-92045-2322https://doaj.org/article/530f2a985a074fb7bdfb89c6da9063f92017-07-01T00:00:00Zhttps://doi.org/10.1038/s41598-017-05728-9https://doaj.org/toc/2045-2322Abstract Multiparametric Magnetic Resonance Imaging (MRI) can provide detailed information of the physical characteristics of rectum tumours. Several investigations suggest that volumetric analyses on anatomical and functional MRI contain clinically valuable information. However, manual delineation of tumours is a time consuming procedure, as it requires a high level of expertise. Here, we evaluate deep learning methods for automatic localization and segmentation of rectal cancers on multiparametric MR imaging. MRI scans (1.5T, T2-weighted, and DWI) of 140 patients with locally advanced rectal cancer were included in our analysis, equally divided between discovery and validation datasets. Two expert radiologists segmented each tumor. A convolutional neural network (CNN) was trained on the multiparametric MRIs of the discovery set to classify each voxel into tumour or non-tumour. On the independent validation dataset, the CNN showed high segmentation accuracy for reader1 (Dice Similarity Coefficient (DSC = 0.68) and reader2 (DSC = 0.70). The area under the curve (AUC) of the resulting probability maps was very high for both readers, AUC = 0.99 (SD = 0.05). Our results demonstrate that deep learning can perform accurate localization and segmentation of rectal cancer in MR imaging in the majority of patients. Deep learning technologies have the potential to improve the speed and accuracy of MRI-based rectum segmentations.Stefano TrebeschiJoost J. M. van GriethuysenDoenja M. J. LambregtsMax J. LahayeChintan ParmarFrans C. H. BakersNicky H. G. M. PetersRegina G. H. Beets-TanHugo J. W. L. AertsNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 7, Iss 1, Pp 1-9 (2017)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Stefano Trebeschi Joost J. M. van Griethuysen Doenja M. J. Lambregts Max J. Lahaye Chintan Parmar Frans C. H. Bakers Nicky H. G. M. Peters Regina G. H. Beets-Tan Hugo J. W. L. Aerts Deep Learning for Fully-Automated Localization and Segmentation of Rectal Cancer on Multiparametric MR
description	Abstract Multiparametric Magnetic Resonance Imaging (MRI) can provide detailed information of the physical characteristics of rectum tumours. Several investigations suggest that volumetric analyses on anatomical and functional MRI contain clinically valuable information. However, manual delineation of tumours is a time consuming procedure, as it requires a high level of expertise. Here, we evaluate deep learning methods for automatic localization and segmentation of rectal cancers on multiparametric MR imaging. MRI scans (1.5T, T2-weighted, and DWI) of 140 patients with locally advanced rectal cancer were included in our analysis, equally divided between discovery and validation datasets. Two expert radiologists segmented each tumor. A convolutional neural network (CNN) was trained on the multiparametric MRIs of the discovery set to classify each voxel into tumour or non-tumour. On the independent validation dataset, the CNN showed high segmentation accuracy for reader1 (Dice Similarity Coefficient (DSC = 0.68) and reader2 (DSC = 0.70). The area under the curve (AUC) of the resulting probability maps was very high for both readers, AUC = 0.99 (SD = 0.05). Our results demonstrate that deep learning can perform accurate localization and segmentation of rectal cancer in MR imaging in the majority of patients. Deep learning technologies have the potential to improve the speed and accuracy of MRI-based rectum segmentations.
format	article
author	Stefano Trebeschi Joost J. M. van Griethuysen Doenja M. J. Lambregts Max J. Lahaye Chintan Parmar Frans C. H. Bakers Nicky H. G. M. Peters Regina G. H. Beets-Tan Hugo J. W. L. Aerts
author_facet	Stefano Trebeschi Joost J. M. van Griethuysen Doenja M. J. Lambregts Max J. Lahaye Chintan Parmar Frans C. H. Bakers Nicky H. G. M. Peters Regina G. H. Beets-Tan Hugo J. W. L. Aerts
author_sort	Stefano Trebeschi
title	Deep Learning for Fully-Automated Localization and Segmentation of Rectal Cancer on Multiparametric MR
title_short	Deep Learning for Fully-Automated Localization and Segmentation of Rectal Cancer on Multiparametric MR
title_full	Deep Learning for Fully-Automated Localization and Segmentation of Rectal Cancer on Multiparametric MR
title_fullStr	Deep Learning for Fully-Automated Localization and Segmentation of Rectal Cancer on Multiparametric MR
title_full_unstemmed	Deep Learning for Fully-Automated Localization and Segmentation of Rectal Cancer on Multiparametric MR
title_sort	deep learning for fully-automated localization and segmentation of rectal cancer on multiparametric mr
publisher	Nature Portfolio
publishDate	2017
url	https://doaj.org/article/530f2a985a074fb7bdfb89c6da9063f9
work_keys_str_mv	AT stefanotrebeschi deeplearningforfullyautomatedlocalizationandsegmentationofrectalcanceronmultiparametricmr AT joostjmvangriethuysen deeplearningforfullyautomatedlocalizationandsegmentationofrectalcanceronmultiparametricmr AT doenjamjlambregts deeplearningforfullyautomatedlocalizationandsegmentationofrectalcanceronmultiparametricmr AT maxjlahaye deeplearningforfullyautomatedlocalizationandsegmentationofrectalcanceronmultiparametricmr AT chintanparmar deeplearningforfullyautomatedlocalizationandsegmentationofrectalcanceronmultiparametricmr AT franschbakers deeplearningforfullyautomatedlocalizationandsegmentationofrectalcanceronmultiparametricmr AT nickyhgmpeters deeplearningforfullyautomatedlocalizationandsegmentationofrectalcanceronmultiparametricmr AT reginaghbeetstan deeplearningforfullyautomatedlocalizationandsegmentationofrectalcanceronmultiparametricmr AT hugojwlaerts deeplearningforfullyautomatedlocalizationandsegmentationofrectalcanceronmultiparametricmr
_version_	1718394278268370944

Deep Learning for Fully-Automated Localization and Segmentation of Rectal Cancer on Multiparametric MR

Ejemplares similares