Global Context and Enhanced Feature Guided Residual Refinement Network for 3D Cardiovascular Image Segmentation

As an important pre-processing step in clinical applications, automatic and accurate 3D cardiovascular image segmentation has attracted more and more attention. However, cardiovascular structures are often with high diversity, blood pool and myocardium shapes are also with large variability, and amb...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Jingjing Liu, Ao Wei, Zhigang Guo, Chang Tang
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Cardiovascular image segmentation deep neural network blood pool and myocardium segmentation Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/fd778ce9befc4f828a66118f59e89ac1
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:fd778ce9befc4f828a66118f59e89ac1
record_format	dspace
spelling	oai:doaj.org-article:fd778ce9befc4f828a66118f59e89ac12021-12-01T00:01:14ZGlobal Context and Enhanced Feature Guided Residual Refinement Network for 3D Cardiovascular Image Segmentation2169-353610.1109/ACCESS.2021.3129333https://doaj.org/article/fd778ce9befc4f828a66118f59e89ac12021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9620094/https://doaj.org/toc/2169-3536As an important pre-processing step in clinical applications, automatic and accurate 3D cardiovascular image segmentation has attracted more and more attention. However, cardiovascular structures are often with high diversity, blood pool and myocardium shapes are also with large variability, and ambiguous cardiac borders make the segmentation task very challenging. In this paper, a novel deep neural network to segment the blood pool and myocardium from three dimensional cardiovascular images is introduced by fully exploiting the global context and complementary information encoded in different feature extraction layers, referred to as GCEFG-R<sup>2</sup>Net briefly. In order to semantically locate the two kinds of regions in a global manner, we design a global context pooling module which can effectively learn context information in a global manner from the deep features extracted from the last two deep layers. Instead of directly using or combining different levels of deep features, we develop an interactive feature aggregation strategy to enhance different levels of deep features by embedding a series of interactive feature aggregation modules. By using the enhanced features, a residual feature refining branch is designed for refining the side outputs in a top-down stream with the guidance of global context features. Finally, the refined side outputs of different layers and the enhanced deep features are combined to generate the final segmentation result by using a feature fusion module. Extensive experiments on two challenge datasets are conducted to demonstrate that the proposed GCEFG-R<sup>2</sup>Net can obtain appealing segmentation results for the blood pool and myocardium and performs better than other state-of-the-art methods.Jingjing LiuAo WeiZhigang GuoChang TangIEEEarticleCardiovascular image segmentationdeep neural networkblood pool and myocardium segmentationElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 155861-155870 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Cardiovascular image segmentation deep neural network blood pool and myocardium segmentation Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Cardiovascular image segmentation deep neural network blood pool and myocardium segmentation Electrical engineering. Electronics. Nuclear engineering TK1-9971 Jingjing Liu Ao Wei Zhigang Guo Chang Tang Global Context and Enhanced Feature Guided Residual Refinement Network for 3D Cardiovascular Image Segmentation
description	As an important pre-processing step in clinical applications, automatic and accurate 3D cardiovascular image segmentation has attracted more and more attention. However, cardiovascular structures are often with high diversity, blood pool and myocardium shapes are also with large variability, and ambiguous cardiac borders make the segmentation task very challenging. In this paper, a novel deep neural network to segment the blood pool and myocardium from three dimensional cardiovascular images is introduced by fully exploiting the global context and complementary information encoded in different feature extraction layers, referred to as GCEFG-R<sup>2</sup>Net briefly. In order to semantically locate the two kinds of regions in a global manner, we design a global context pooling module which can effectively learn context information in a global manner from the deep features extracted from the last two deep layers. Instead of directly using or combining different levels of deep features, we develop an interactive feature aggregation strategy to enhance different levels of deep features by embedding a series of interactive feature aggregation modules. By using the enhanced features, a residual feature refining branch is designed for refining the side outputs in a top-down stream with the guidance of global context features. Finally, the refined side outputs of different layers and the enhanced deep features are combined to generate the final segmentation result by using a feature fusion module. Extensive experiments on two challenge datasets are conducted to demonstrate that the proposed GCEFG-R<sup>2</sup>Net can obtain appealing segmentation results for the blood pool and myocardium and performs better than other state-of-the-art methods.
format	article
author	Jingjing Liu Ao Wei Zhigang Guo Chang Tang
author_facet	Jingjing Liu Ao Wei Zhigang Guo Chang Tang
author_sort	Jingjing Liu
title	Global Context and Enhanced Feature Guided Residual Refinement Network for 3D Cardiovascular Image Segmentation
title_short	Global Context and Enhanced Feature Guided Residual Refinement Network for 3D Cardiovascular Image Segmentation
title_full	Global Context and Enhanced Feature Guided Residual Refinement Network for 3D Cardiovascular Image Segmentation
title_fullStr	Global Context and Enhanced Feature Guided Residual Refinement Network for 3D Cardiovascular Image Segmentation
title_full_unstemmed	Global Context and Enhanced Feature Guided Residual Refinement Network for 3D Cardiovascular Image Segmentation
title_sort	global context and enhanced feature guided residual refinement network for 3d cardiovascular image segmentation
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/fd778ce9befc4f828a66118f59e89ac1
work_keys_str_mv	AT jingjingliu globalcontextandenhancedfeatureguidedresidualrefinementnetworkfor3dcardiovascularimagesegmentation AT aowei globalcontextandenhancedfeatureguidedresidualrefinementnetworkfor3dcardiovascularimagesegmentation AT zhigangguo globalcontextandenhancedfeatureguidedresidualrefinementnetworkfor3dcardiovascularimagesegmentation AT changtang globalcontextandenhancedfeatureguidedresidualrefinementnetworkfor3dcardiovascularimagesegmentation
_version_	1718406181017354240

Global Context and Enhanced Feature Guided Residual Refinement Network for 3D Cardiovascular Image Segmentation

Ejemplares similares