Late Gadolinium Enhanced (LGE) Magnetic Resonance Imaging (MRI) is a new emerging non-invasive technique which might be employed for the non-invasive quantification of left atrium (LA) myocardial fibrotic tissue in patients affected by atrial fibrillation. Nowadays, the analysis of LGE MRI relies on manual tracing of LA boundaries. An automated LA segmentation approach for the quantification of scar tissue would be highly desirable. This study focuses on the design of a fully automated LGE MRI segmentation pipeline which includes a convolutional neural network (CNN) based on the successful architecture U-Net. The CNN was trained with the data available from the Statistical Atlases and Computational Modelling of the Heart 2018 Atrial Segmentation Challenge (100 cardiac data) with two different approaches: using both stacks of 2-D axial slices and using 3-D data. Mean Dice coefficients on the test set were 0.896 and 0.914 by using the 2-D and 3-D approaches, respectively. Contour accuracy was highly variable along the LA longitudinal axis showing poorest results in correspondence of the pulmonary veins. These results suggest that, despite the increase of the number of trainable parameters, the proposed 3-D CNN learns better features leading to higher performance, feasible for a real clinical application.

An Automated Approach Based on a Convolutional Neural Network for Left Atrium Segmentation from Late Gadolinium Enhanced Magnetic Resonance Imaging

Borra D.;Fabbri C.;Masci A.;Corsi C.
2019

Abstract

Late Gadolinium Enhanced (LGE) Magnetic Resonance Imaging (MRI) is a new emerging non-invasive technique which might be employed for the non-invasive quantification of left atrium (LA) myocardial fibrotic tissue in patients affected by atrial fibrillation. Nowadays, the analysis of LGE MRI relies on manual tracing of LA boundaries. An automated LA segmentation approach for the quantification of scar tissue would be highly desirable. This study focuses on the design of a fully automated LGE MRI segmentation pipeline which includes a convolutional neural network (CNN) based on the successful architecture U-Net. The CNN was trained with the data available from the Statistical Atlases and Computational Modelling of the Heart 2018 Atrial Segmentation Challenge (100 cardiac data) with two different approaches: using both stacks of 2-D axial slices and using 3-D data. Mean Dice coefficients on the test set were 0.896 and 0.914 by using the 2-D and 3-D approaches, respectively. Contour accuracy was highly variable along the LA longitudinal axis showing poorest results in correspondence of the pulmonary veins. These results suggest that, despite the increase of the number of trainable parameters, the proposed 3-D CNN learns better features leading to higher performance, feasible for a real clinical application.
2019
Computing in Cardiology
1
4
Borra D.; Fabbri C.; Masci A.; Esposito L.; Andalo A.; Corsi C.
File in questo prodotto:
Eventuali allegati, non sono esposti

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11585/811044
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact