IEICE Technical Committee Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
Tech. Rep. Archives
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2019-01-23 09:45
Feature Selection from Imbalanced Data -- Pathological Pattern Classification in Endocytoscopic Images --
Hayato Itoh (Nagoya Univ.), Yuichi Mori, Masashi Misawa (Showa Univ.), Masahiro Oda (Nagoya Univ.), Shin-Ei Kudo (Showa Univ.), Kensaku Mori (Nagoya Univ.) MI2018-87
Abstract (in Japanese) (See Japanese page) 
(in English) Endocytoscope gives ultramagnified observation that enables physicians to achieve minimally invasive and real-time diagnosis in colonoscopy. Since this modality is a quite new, a pathological image classifier is required for the support of non-expert physicians. In addition to the ununiformity of occurrence frequency of pathological
patterns, data sampling by physicians includes bias. We have to handle imbalances of data to design an accurate pathological classifier. We propose feature-selection method that selects discriminative feature from imbalanced data for training of pathological classifier. We experimentally evaluated the proposed method by comparing the classification accuracy between before and after feature selection with about 50,000 endocytoscopic images. Our method achieves 2.7% improvement of classification accuracy with more accurate likelihood estimation than original texture features.
Keyword (in Japanese) (See Japanese page) 
(in English) Endocytoscopy / automated pathological diagnosis / feature selection / texture feature / manifold learning / definite canonicalisation / /  
Reference Info. IEICE Tech. Rep., vol. 118, no. 412, MI2018-87, pp. 109-114, Jan. 2019.
Paper # MI2018-87 
Date of Issue 2019-01-15 (MI) 
ISSN 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 MI2018-87

Conference Information
Committee MI  
Conference Date 2019-01-22 - 2019-01-23 
Place (in Japanese) (See Japanese page) 
Place (in English)  
Topics (in Japanese) (See Japanese page) 
Topics (in English) Medical Image Engineering, Analysis, Recognition, etc. 
Paper Information
Registration To MI 
Conference Code 2019-01-MI 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Feature Selection from Imbalanced Data 
Sub Title (in English) Pathological Pattern Classification in Endocytoscopic Images 
Keyword(1) Endocytoscopy  
Keyword(2) automated pathological diagnosis  
Keyword(3) feature selection  
Keyword(4) texture feature  
Keyword(5) manifold learning  
Keyword(6) definite canonicalisation  
Keyword(7)  
Keyword(8)  
1st Author's Name Hayato Itoh  
1st Author's Affiliation Nagoya University (Nagoya Univ.)
2nd Author's Name Yuichi Mori  
2nd Author's Affiliation Showa University (Showa Univ.)
3rd Author's Name Masashi Misawa  
3rd Author's Affiliation Showa University (Showa Univ.)
4th Author's Name Masahiro Oda  
4th Author's Affiliation Nagoya University (Nagoya Univ.)
5th Author's Name Shin-Ei Kudo  
5th Author's Affiliation Showa University (Showa Univ.)
6th Author's Name Kensaku Mori  
6th Author's Affiliation Nagoya University (Nagoya Univ.)
7th Author's Name  
7th Author's Affiliation ()
8th Author's Name  
8th Author's Affiliation ()
9th Author's Name  
9th Author's Affiliation ()
10th Author's Name  
10th Author's Affiliation ()
11th Author's Name  
11th Author's Affiliation ()
12th Author's Name  
12th Author's Affiliation ()
13th Author's Name  
13th Author's Affiliation ()
14th Author's Name  
14th Author's Affiliation ()
15th Author's Name  
15th Author's Affiliation ()
16th Author's Name  
16th Author's Affiliation ()
17th Author's Name  
17th Author's Affiliation ()
18th Author's Name  
18th Author's Affiliation ()
19th Author's Name  
19th Author's Affiliation ()
20th Author's Name  
20th Author's Affiliation ()
Speaker Author-1 
Date Time 2019-01-23 09:45:00 
Presentation Time 15 minutes 
Registration for MI 
Paper # MI2018-87 
Volume (vol) vol.118 
Number (no) no.412 
Page pp.109-114 
#Pages
Date of Issue 2019-01-15 (MI) 


[Return to Top Page]

[Return to IEICE Web Page]


The Institute of Electronics, Information and Communication Engineers (IEICE), Japan