FOANet: A Focus of Attention Network with Application to Myocardium Segmentation
From LRDE
- Authors
- Zhou Zhao, Nicolas Boutry, Élodie Puybareau, Thierry Géraud
- Where
- Proceedings of the 25th International Conference on Pattern Recognition (ICPR)
- Place
- Milan, Italy
- Type
- inproceedings
- Publisher
- IEEE
- Projects
- Olena
- Keywords
- Image
- Date
- 2020-11-02
Abstract
In myocardium segmentation of cardiac magnetic resonance images, ambiguities often appear near the boundaries of the target domains due to tissue similarities. To address this issue, we propose a new architecture, called FOANet, which can be decomposed in three main steps: a localization stepa Gaussian-based contrast enhancement step, and a segmentation step. This architecture is supplied with a hybrid loss function that guides the FOANet to study the transformation relationship between the input image and the corresponding label in a three-level hierarchy (pixel-patch- and map-level), which is helpful to improve segmentation and recovery of the boundaries. We demonstrate the efficiency of our approach on two public datasets in terms of regional and boundary segmentations.
Documents
Bibtex (lrde.bib)
@InProceedings{ zhao.20.icpr.1, author = {Zhou Zhao and Nicolas Boutry and \'Elodie Puybareau and Thierry G\'eraud}, title = {{FOANet}: {A} Focus of Attention Network with Application to Myocardium Segmentation}, booktitle = {Proceedings of the 25th International Conference on Pattern Recognition (ICPR)}, year = 2021, pages = {1120--1127}, month = jan, address = {Milan, Italy}, publisher = {IEEE}, abstract = {In myocardium segmentation of cardiac magnetic resonance images, ambiguities often appear near the boundaries of the target domains due to tissue similarities. To address this issue, we propose a new architecture, called FOANet, which can be decomposed in three main steps: a localization step, a Gaussian-based contrast enhancement step, and a segmentation step. This architecture is supplied with a hybrid loss function that guides the FOANet to study the transformation relationship between the input image and the corresponding label in a three-level hierarchy (pixel-, patch- and map-level), which is helpful to improve segmentation and recovery of the boundaries. We demonstrate the efficiency of our approach on two public datasets in terms of regional and boundary segmentations.}, doi = {10.1109/ICPR48806.2021.9412016} }