| Paper Abstract and Keywords |
| Presentation |
2010-01-28 10:05
A Strategy to Improve PET Neuroreceptor Analysis MA1 Based on Statistics of Measurement Errors Naoki Hoshino, Hidekata Hontani (Nagoya Inst. of Tech.), Mika Naganawa, Kazuya Sakaguchi (National Inst. of Radiological Sci.), Muneyuki Sakata, Kiichi Ishiwata (Tokyo Metro Inst. of Gerontology), Yuichi Kimura (National Inst. of Radiological Sci.) MI2009-76 |
| Abstract |
(in Japanese) |
(See Japanese page) |
| (in English) |
In this article, we propose a new method for improving MA1, which
estimates total distribution volume $V_{\rm T}$, the parameter
reflecting a neuroreceptor density. MA1 applies a linear regression
analysis to a set of data computed from the tissue time activity curve
(tTAC) and the plasma time activity curve (pTAC) to obtain the value
of $V_{\rm T}$. The problem is that the linear regression is not
supported by the statistics of the noise included in the data. Hence,
the proposed method de-noises measured tTAC to estimate the
corresponding true tTAC by means of MAP estimation. For the MAP
estimation, we need the prior distribution of tTAC and the likelihood
one on the measured tTAC. The proposed method estimates the former one
based on simulations and estimates the latter one from the
measurements. We evaluated the performance of MA1 and of the proposed
method by using simulated tTACs and by using clinical data considering
$^{11}$C-SA4503 that is a probe of $\sigma_{1}$ receptors. The
evaluation results show that our proposed method improves the
variances of the estimates. |
| Keyword |
(in Japanese) |
(See Japanese page) |
| (in English) |
PET / total distribution volume / MA1 / MAP estimation / denoising / kinetic analysis / / |
| Reference Info. |
IEICE Tech. Rep., vol. 109, no. 407, MI2009-76, pp. 7-12, Jan. 2010. |
| Paper # |
MI2009-76 |
| Date of Issue |
2010-01-21 (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 |
MI2009-76 |