Paper Abstract and Keywords |
Presentation |
2017-05-26 09:20
An Age Estimation Method Using Brain Local Features of Brain MRI Images and Its Performance Evaluation Ryuichi Fujimoto, Koichi Ito (Tohoku Univ.), Kai Wu (South China Univ.), Kazunori Sato, Yasuyuki Taki (Tohoku Univ.), Hiroshi Fukuda (Tohoku Medical and Pharmaceutical Univ.), Takafumi Aoki (Tohoku Univ.) SIP2017-13 IE2017-13 PRMU2017-13 MI2017-13 |
Abstract |
(in Japanese) |
(See Japanese page) |
(in English) |
It is known that brain tissues have age-related morphological changes through a set of statistical analysis
using large-scale brain MRI image databases.
This fact allows us to estimate the age of a subject from brain
MRI images by evaluating brain morphological changes with healthy aging.
The age estimated from morphological changes of a human brain
can be used for diagnostic support and early identification
of brain disorders such as Alzheimer's Disease.
This paper proposes an age estimation method using local features
extracted from T1-weighted MRI images.
Local features are defined by regional volume calculated from 1,024 local
regions of GM, WM and CSF.
We also add cortical thickness parcellated by Destrieux atlas to
brain local features in order to improve the accuracy of
age estimation.
We evaluate performance of the proposed method using a large-scale dataset of healty Japanese. |
Keyword |
(in Japanese) |
(See Japanese page) |
(in English) |
MRI / T1-weighted image / age estimation / brain aging / local features / machine learning / relevance vector machine / |
Reference Info. |
IEICE Tech. Rep., vol. 117, no. 50, MI2017-13, pp. 67-70, May 2017. |
Paper # |
MI2017-13 |
Date of Issue |
2017-05-18 (SIP, IE, PRMU, 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 |
SIP2017-13 IE2017-13 PRMU2017-13 MI2017-13 |
|