Automatic segmentation with detection of local segmentation failures in cardiac MRI

Abstract Segmentation of cardiac anatomical structures in cardiac magnetic resonance images (CMRI) is a prerequisite for automatic diagnosis and prognosis of cardiovascular diseases. To increase robustness and performance of segmentation methods this study combines automatic segmentation and assessm...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Jörg Sander, Bob D. de Vos, Ivana Išgum
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2020
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/d0ad7ca897184f47b52c544724f78830
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:d0ad7ca897184f47b52c544724f78830
record_format	dspace
spelling	oai:doaj.org-article:d0ad7ca897184f47b52c544724f788302021-12-02T16:18:03ZAutomatic segmentation with detection of local segmentation failures in cardiac MRI10.1038/s41598-020-77733-42045-2322https://doaj.org/article/d0ad7ca897184f47b52c544724f788302020-12-01T00:00:00Zhttps://doi.org/10.1038/s41598-020-77733-4https://doaj.org/toc/2045-2322Abstract Segmentation of cardiac anatomical structures in cardiac magnetic resonance images (CMRI) is a prerequisite for automatic diagnosis and prognosis of cardiovascular diseases. To increase robustness and performance of segmentation methods this study combines automatic segmentation and assessment of segmentation uncertainty in CMRI to detect image regions containing local segmentation failures. Three existing state-of-the-art convolutional neural networks (CNN) were trained to automatically segment cardiac anatomical structures and obtain two measures of predictive uncertainty: entropy and a measure derived by MC-dropout. Thereafter, using the uncertainties another CNN was trained to detect local segmentation failures that potentially need correction by an expert. Finally, manual correction of the detected regions was simulated in the complete set of scans of 100 patients and manually performed in a random subset of scans of 50 patients. Using publicly available CMR scans from the MICCAI 2017 ACDC challenge, the impact of CNN architecture and loss function for segmentation, and the uncertainty measure was investigated. Performance was evaluated using the Dice coefficient, 3D Hausdorff distance and clinical metrics between manual and (corrected) automatic segmentation. The experiments reveal that combining automatic segmentation with manual correction of detected segmentation failures results in improved segmentation and to 10-fold reduction of expert time compared to manual expert segmentation.Jörg SanderBob D. de VosIvana IšgumNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 10, Iss 1, Pp 1-19 (2020)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Jörg Sander Bob D. de Vos Ivana Išgum Automatic segmentation with detection of local segmentation failures in cardiac MRI
description	Abstract Segmentation of cardiac anatomical structures in cardiac magnetic resonance images (CMRI) is a prerequisite for automatic diagnosis and prognosis of cardiovascular diseases. To increase robustness and performance of segmentation methods this study combines automatic segmentation and assessment of segmentation uncertainty in CMRI to detect image regions containing local segmentation failures. Three existing state-of-the-art convolutional neural networks (CNN) were trained to automatically segment cardiac anatomical structures and obtain two measures of predictive uncertainty: entropy and a measure derived by MC-dropout. Thereafter, using the uncertainties another CNN was trained to detect local segmentation failures that potentially need correction by an expert. Finally, manual correction of the detected regions was simulated in the complete set of scans of 100 patients and manually performed in a random subset of scans of 50 patients. Using publicly available CMR scans from the MICCAI 2017 ACDC challenge, the impact of CNN architecture and loss function for segmentation, and the uncertainty measure was investigated. Performance was evaluated using the Dice coefficient, 3D Hausdorff distance and clinical metrics between manual and (corrected) automatic segmentation. The experiments reveal that combining automatic segmentation with manual correction of detected segmentation failures results in improved segmentation and to 10-fold reduction of expert time compared to manual expert segmentation.
format	article
author	Jörg Sander Bob D. de Vos Ivana Išgum
author_facet	Jörg Sander Bob D. de Vos Ivana Išgum
author_sort	Jörg Sander
title	Automatic segmentation with detection of local segmentation failures in cardiac MRI
title_short	Automatic segmentation with detection of local segmentation failures in cardiac MRI
title_full	Automatic segmentation with detection of local segmentation failures in cardiac MRI
title_fullStr	Automatic segmentation with detection of local segmentation failures in cardiac MRI
title_full_unstemmed	Automatic segmentation with detection of local segmentation failures in cardiac MRI
title_sort	automatic segmentation with detection of local segmentation failures in cardiac mri
publisher	Nature Portfolio
publishDate	2020
url	https://doaj.org/article/d0ad7ca897184f47b52c544724f78830
work_keys_str_mv	AT jorgsander automaticsegmentationwithdetectionoflocalsegmentationfailuresincardiacmri AT bobddevos automaticsegmentationwithdetectionoflocalsegmentationfailuresincardiacmri AT ivanaisgum automaticsegmentationwithdetectionoflocalsegmentationfailuresincardiacmri
_version_	1718384200839593984

Automatic segmentation with detection of local segmentation failures in cardiac MRI

Ejemplares similares