Stacked and parallel U-nets with multi-output for myocardial pathology segmentation

Authors

Zhou Zhao, Nicolas Boutry, Elodie Puybareau

Where

Myocardial Pathology Segmentation Combining Multi-Sequence CMR Challenge

Type

inproceedings

Date

2020-01-01

Abstract

In the field of medical imaging, many different image modalities contain different information, helping practitionners to make diagnostic, follow-up, etc. To better analyze images, mixing multi-modalities information has become a trend. This paper provides one cascaded UNet framework and uses three different modalities (the late gadolinium enhancement (LGE) CMR sequence, the balanced- Steady State Free Precession (bSSFP) cine sequence and the T2-weighted CMR) to complete the segmentation of the myocardium, scar and edema in the context of the MICCAI 2020 myocardial pathology segmentation combining multi-sequence CMR Challenge dataset (MyoPS 2020). We evaluate the proposed method with 5-fold-cross-validation on the MyoPS 2020 dataset.

Documents

• Paper

Bibtex (lrde.bib)

@InProceedings{	  zhao.19.myops,
  title		= {Stacked and parallel {U}-nets with multi-output for
		  myocardial pathology segmentation},
  author	= {Zhou Zhao and Nicolas Boutry and Elodie Puybareau},
  booktitle	= {Myocardial Pathology Segmentation Combining Multi-Sequence
		  CMR Challenge},
  pages		= {138--145},
  year		= {2020},
  organization	= {Springer},
  volume	= {12554},
  series	= {Lecture Notes in Computer Science},
  doi		= {10.1007/978-3-030-65651-5_13},
  abstract	= {In the field of medical imaging, many different image
		  modalities contain different information, helping
		  practitionners to make diagnostic, follow-up, etc. To
		  better analyze images, mixing multi-modalities information
		  has become a trend. This paper provides one cascaded UNet
		  framework and uses three different modalities (the late
		  gadolinium enhancement (LGE) CMR sequence, the balanced-
		  Steady State Free Precession (bSSFP) cine sequence and the
		  T2-weighted CMR) to complete the segmentation of the
		  myocardium, scar and edema in the context of the MICCAI
		  2020 myocardial pathology segmentation combining
		  multi-sequence CMR Challenge dataset (MyoPS 2020). We
		  evaluate the proposed method with 5-fold-cross-validation
		  on the MyoPS 2020 dataset.}
}