Abstract #3136
Improved T2* assessment in liver iron overload by 2D fuzzy c-mean clustering
Pairash Saiviroonporn 1 , Vip Viprakasit 2 , Rungroj Krittayaphong 3 , and John C Wood 4
1
Radiology Department, Faculty of Medicine
Siriraj Hospital, Mahidol University, Bangkoknoi,
Bangkok, Thailand,
2
Department
of Pediatrics, Faculty of Medicine Siriraj Hospital,
Mahidol University, Bangkoknoi, Bangkok, Thailand,
3
Department
of Medicine, Faculty of Medicine Siriraj Hospital,
Mahidol University, Bangkoknoi, Bangkok, Thailand,
4
Department
of Pediatrics, Childrens Hospital Los Angeles, Keck
School of Medicine, University of Southern California,
Los Angeles, California, United States
The study investigated the usefulness of the 2D fuzzy
c-mean (FCM) clustering to lower the variability of the
T2* liver iron assessment by separated the vessel pixels
from parenchyma. The manual and 2D-FCM segmentations
were performed on the multi-echo T2* images and their
LIC maps of 60 thalassemia major patients. The 2D FCM
method can correctly segment the parenchyma and vessel
pixels by 95.77.9% and 99.52.4%, respectively. The
variability of the T2* measurement then can be lower by
32%, but finding the optimal clustering variables are
necessary before it can be practically employed.
This abstract and the presentation materials are available to members only;
a login is required.
Join Here
