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 2017-11-06 10:40
On the influence of Dice loss function in multi-class organ segmentation of abdominal CT using 3D fully convolutional networks
Chen Shen, Holger R. Roth, Hirohisa Oda, Masahiro Oda, Yuichiro Hayashi (Nagoya Univ.), Kazunari Misawa (Aichi Cancer Central Center Hospital), Kensaku Mori (Nagoya Univ.) MICT2017-29 MI2017-51
Abstract (in Japanese) (See Japanese page) 
(in English) Deep learning-based methods achieved impressive results in segmentations from medical images. With the development of 3D fully convolutional networks (FCNs), it has become feasible to produce improved results for multi-organ segmentation of 3D computed tomography (CT) images. The results of multi-organ segmentation using deep learning-based methods not only depend on the choice of networks architecture, but also strongly rely on the choice of loss function. In this paper, we present a discussion on the influence of Dice-based loss functions for multi-class organ segmentation using a dataset of abdominal CT volumes. We investigated three different types of weighting the Dice loss functions based on class label frequencies (uniform, simple and square) and evaluate their influence on segmentation accuracies. Furthermore, we compared the influence of different initial learning rates. We achieved average Dice scores of 81.3%, 59.5% and 31.7% for uniform, simple and square types of weighting when the learning rate is 0.001, and 78.2%, 81.0% and 58.5% for each weighting when the learning rate is 0.01. Our experiments indicated a strong relationship between class balancing weights and initial learning rate in training.
Keyword (in Japanese) (See Japanese page) 
(in English) multi-organ segmentation / Dice loss function / fully convolutional network / computed tomography / / / /  
Reference Info. IEICE Tech. Rep., vol. 117, no. 281, MI2017-51, pp. 15-20, Nov. 2017.
Paper # MI2017-51 
Date of Issue 2017-10-30 (MICT, 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 MICT2017-29 MI2017-51

Conference Information
Committee MI MICT  
Conference Date 2017-11-06 - 2017-11-06 
Place (in Japanese) (See Japanese page) 
Place (in English) Sunport Hall Takamatsu 
Topics (in Japanese) (See Japanese page) 
Topics (in English) Medical Imaging, etc. 
Paper Information
Registration To MI 
Conference Code 2017-11-MI-MICT 
Language English (Japanese title is available) 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) On the influence of Dice loss function in multi-class organ segmentation of abdominal CT using 3D fully convolutional networks 
Sub Title (in English)  
Keyword(1) multi-organ segmentation  
Keyword(2) Dice loss function  
Keyword(3) fully convolutional network  
Keyword(4) computed tomography  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Chen Shen  
1st Author's Affiliation Nagoya University (Nagoya Univ.)
2nd Author's Name Holger R. Roth  
2nd Author's Affiliation Nagoya University (Nagoya Univ.)
3rd Author's Name Hirohisa Oda  
3rd Author's Affiliation Nagoya University (Nagoya Univ.)
4th Author's Name Masahiro Oda  
4th Author's Affiliation Nagoya University (Nagoya Univ.)
5th Author's Name Yuichiro Hayashi  
5th Author's Affiliation Nagoya University (Nagoya Univ.)
6th Author's Name Kazunari Misawa  
6th Author's Affiliation Aichi Cancer Central Center Hospital (Aichi Cancer Central Center Hospital)
7th Author's Name Kensaku Mori  
7th Author's Affiliation Nagoya University (Nagoya Univ.)
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 2017-11-06 10:40:00 
Presentation Time 20 minutes 
Registration for MI 
Paper # MICT2017-29, MI2017-51 
Volume (vol) vol.117 
Number (no) no.280(MICT), no.281(MI) 
Page pp.15-20 
#Pages
Date of Issue 2017-10-30 (MICT, MI) 


[Return to Top Page]

[Return to IEICE Web Page]


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