Paper Abstract and Keywords |
Presentation |
2011-01-20 11:30
Brain Shape Normalization in Neonatal MR Images Aya Hashioka (Univ. of Hyogo), Syoji Kobashi, Kei Kuramoto (Univ. of Hyogo/ WPI-IFReC), Yuki Wakata, Kumiko Ando, Reiichi Ishikura (Hyogo College of Medicine), Tomomoto Ishikawa (Ishikawa Hospital), Shozo Hirota (Hyogo College of Medicine), Yutaka Hata (Univ. of Hyogo/ WPI-IFReC) MI2010-103 |
Abstract |
(in Japanese) |
(See Japanese page) |
(in English) |
In order to diagnose the cerebral diseases in the early stage, diagnosing method for neonatal brain must be studied. Voxel-based morphometry (VBM) is an approach to diagnose the cerebral diseases using MR (magnetic resonance) images. VBM normalizes the evaluating brain MR images into a template, and evaluates the distribution of gray matter density statistically. Because the template is produced from adult brain MR images, it is difficult to apply the conventional normalization to the neonatal brain. In this article, three new normalization approaches for the neonatal brain are proposed. The 1st approach defines an individual neonatal brain as a template. The 2nd approach introduces skull stripping into the 1st approach. The 3rd approach produces a neonatal template brain from some neonatal brains. Normalization accuracy was evaluated by using mutual information, position of anterior and posterior commissures (AC and PC) and angle of AC-PC line. The results in 14 neonates (revised age between -4 and 4 weeks) showed that the normalization accuracy of the 3rd approach was significantly higher than the conventional normalization and the other two approaches. |
Keyword |
(in Japanese) |
(See Japanese page) |
(in English) |
voxel-based morphometry / MR image / neonate / brain / normalization / mutual information / anterior commissure / posterior commisure |
Reference Info. |
IEICE Tech. Rep., vol. 110, no. 364, MI2010-103, pp. 115-120, Jan. 2011. |
Paper # |
MI2010-103 |
Date of Issue |
2011-01-12 (MI) |
ISSN |
Print edition: ISSN 0913-5685 Online edition: ISSN 2432-6380 |
Copyright and reproduction |
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
Download PDF |
MI2010-103 |
|