| Paper Abstract and Keywords |
| Presentation |
2014-11-13 13:20
A Hierarchical Type Segmentation Algorithm based on Support Vector Machine for Computer-Aided Diagnosis of Colorectal Endoscopic Images Takumi Okamoto, Tetsushi Koide, Anh-Tuan Hoang, Koki Sugi, Tatsuya Shimizu, Toru Tamaki, Bisser Raytchev, Kazufumi Kaneda, Yoko Kominami, Shigeto Yoshida, Shinji Tanaka (Hiroshima Univ.) CPSY2014-56 |
| Abstract |
(in Japanese) |
(See Japanese page) |
| (in English) |
With the increase of colorectal cancer patients in recent years, the needs of quantitative evaluation of colorectal cancer are increased, and the Computer-Aided Diagnosis (CAD) system which supports doctor's diagnosis is essential. This paper describes type identification module in the above CAD system for colorectal endoscopic images with narrow band imaging (NBI) magnification. In type identification module, the raster scan of a colorectal endoscopic image with Scan Window (SW), is carried out and the type identification of each SW is performed. In this paper, we propose a hierarchical type segmentation algorithm which realizes improvement in accuracy and resolution with combination of different SW sizes. The simulation results are also presented for the cases in which the feature extraction is performed in dense and rough for small and large scan window sizes, respectively. |
| Keyword |
(in Japanese) |
(See Japanese page) |
| (in English) |
Support Vector Mahine (SVM) / Colorectal Endoscopic Images / Computer-Aided Diagnosis (CAD) / Hierarchical Type Segmentation / FPGA / / / |
| Reference Info. |
IEICE Tech. Rep., vol. 114, no. 302, CPSY2014-56, pp. 13-18, Nov. 2014. |
| Paper # |
CPSY2014-56 |
| Date of Issue |
2014-11-06 (CPSY) |
| 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 |
CPSY2014-56 |