2D-3D Fully convolutional neural networks for cardiac MR segmentation

File Description SizeFormat 
1707.09813v1.pdfAccepted version659.62 kBAdobe PDFView/Open
Title: 2D-3D Fully convolutional neural networks for cardiac MR segmentation
Authors: Patravali, J
Jain, S
Chilamkurthy, S
Item Type: Conference Paper
Abstract: In this paper, we develop a 2D and 3D segmentation pipelines for fully automated cardiac MR image segmentation using Deep Convolutional Neural Networks (CNN). Our models are trained end-to-end from scratch using the ACD Challenge 2017 dataset comprising of 100 studies, each containing Cardiac MR images in End Diastole and End Systole phase. We show that both our segmentation models achieve near state-of-the-art performance scores in terms of distance metrics and have convincing accuracy in terms of clinical parameters. A comparative analysis is provided by introducing a novel dice loss function and its combination with cross entropy loss. By exploring different network structures and comprehensive experiments, we discuss several key insights to obtain optimal model performance, which also is central to the theme of this challenge.
Issue Date: 15-Mar-2019
Date of Acceptance: 10-Sep-2017
URI: http://hdl.handle.net/10044/1/71234
DOI: https://doi.org/10.1007/978-3-319-75541-0_14
ISBN: 9783319755403
Publisher: Springer
Start Page: 130
End Page: 139
Journal / Book Title: Statistical Atlases and Computational Models of the Heart. ACDC and MMWHS Challenges
Volume: 10663
Copyright Statement: © 2018 Springer International Publishing AG.
Conference Name: Statistical Atlases and Computational Models of the Heart. ACDC and MMWHS Challenges
Keywords: cs.CV
cs.CV
cs.CV
cs.CV
Notes: Accepted in STACOM '17
Start Date: 2017-09-10
Finish Date: 2017-09-14
Conference Place: Quebec City, Canada
Online Publication Date: 2018-03-15
Appears in Collections:Faculty of Engineering
Computing



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commonsx