eprintid: 2771
rev_number: 8
eprint_status: archive
userid: 69
dir: disk0/00/00/27/71
datestamp: 2015-10-13 08:07:33
lastmod: 2015-10-13 08:08:43
status_changed: 2015-10-13 08:07:33
type: book_section
metadata_visibility: show
creators_name: Mukhopadhyay, Anirban
creators_name: Oksuz, Ilkay
creators_name: Bevilacqua, Marco
creators_name: Dharmakumar, Rohan
creators_name: Tsaftaris, Sotirios A.
creators_id: 
creators_id: ilkay.oksuz@imtlucca.it
creators_id: 
creators_id: 
creators_id: 
title: Unsupervised Myocardial Segmentation for Cardiac MRI
ispublished: pub
subjects: QA75
subjects: QH301
subjects: QM
divisions: CSA
full_text_status: none
keywords: Unsupervised Segmentation; Dictionary Learning; BOLD; CINE; MRI
abstract: Though unsupervised segmentation was a de-facto standard for cardiac MRI segmentation early on, recently cardiac MRI segmentation literature has favored fully supervised techniques such as Dictionary Learning and Atlas-based techniques. But, the benefits of unsupervised techniques e.g., no need for large amount of training data and better potential of handling variability in anatomy and image contrast, is more evident with emerging cardiac MR modalities. For example, CP-BOLD is a new MRI technique that has been shown to detect ischemia without any contrast at stress but also at rest conditions. Although CP-BOLD looks similar to standard CINE, changes in myocardial intensity patterns and shape across cardiac phases, due to the heart’s motion, BOLD effect and artifacts affect the underlying mechanisms of fully supervised segmentation techniques resulting in a significant drop in segmentation accuracy. In this paper, we present a fully unsupervised technique for segmenting myocardium from the background in both standard CINE MR and CP-BOLD MR. We combine appearance with motion information (obtained via Optical Flow) in a dictionary learning framework to sparsely represent important features in a low dimensional space and separate myocardium from background accordingly. Our fully automated method learns background-only models and one class classifier provides myocardial segmentation. The advantages of the proposed technique are demonstrated on a dataset containing CP-BOLD MR and standard CINE MR image sequences acquired in baseline and ischemic condition across 10 canine subjects, where our method outperforms state-of-the-art supervised segmentation techniques in CP-BOLD MR and performs at-par for standard CINE MR.
date: 2015
date_type: published
series: Lecture Notes in Computer Science
volume: 9351
publisher: Springer
pagerange: 12-20
pages: 9
id_number: 10.1007/978-3-319-24574-4_2
refereed: TRUE
isbn: 978-3-319-24573-7
book_title: Medical Image Computing and Computer-Assisted Intervention. MICCAI 2015
official_url: http://link.springer.com/chapter/10.1007/978-3-319-24574-4_2#
citation:   Mukhopadhyay, Anirban and Oksuz, Ilkay and Bevilacqua, Marco and Dharmakumar, Rohan and Tsaftaris, Sotirios A.  Unsupervised Myocardial Segmentation for Cardiac MRI.   In:  Medical Image Computing and Computer-Assisted Intervention. MICCAI 2015.   Lecture Notes in Computer Science, 9351 .  Springer, pp. 12-20.  ISBN 978-3-319-24573-7      (2015)