Information: Join today and make your research activities more affordable! Technical workshop participation fees and annual registration fees are available at member rates.
Notice: [Important] Announcement of Changes to Registration Fee Payment and Manuscript Upload Procedures for IEICE Technical Meetings
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 2026-03-06 12:59
A Study on Masking Ratio and Dataset Expansion for Pre-Training Laparoscopic Image Foundation Models
Kaede Yasuda, Yuichiro Hayashi, Masahiro Oda (Nagoya Univ.), Kensaku Mori (Nagoya Univ./NII/) MI2025-101
Abstract (in Japanese) (See Japanese page) 
(in English) In this paper, we propose a method for pre-training a Vision Transformer (ViT) to create a foundation model applicable to multiple laparoscopic image recognition tasks. Currently, laparoscopic surgery is widely adopted due to its ability to reduce physical burden on patients. However, it imposes a significant load on surgeons, highlighting the need for computer-aided support. Since the shortage of labeled data is a major challenge in medical image analysis, Self-Supervised Learning (SSL), which leverages
large-scale unlabeled data, has attracted attention. In this study, we improved EndoViT, an existing pre-training model based on Endo700k, by (1) modifying the masking strategy and (2) performing dataset expansion with our original large-scale dataset. Quantitative evaluations on three downstream tasks, Semantic Segmentation, Action Triplet Detection (ATD), and Surgical Phase Recognition (SPR), demonstrated that the proposed method improved mIoU by approximately 4 points in Semantic Segmentation.
On the other hand, the results indicated that the conventional masking strategy remains superior for ATD and SPR.
Keyword (in Japanese) (See Japanese page) 
(in English) Laparoscopic Surgery / Vision Transformer / Masked Autoencoder / Foundation Model / Surgical Scene Recognition / / /  
Reference Info. IEICE Tech. Rep., vol. 125, no. 395, MI2025-101, pp. 155-158, March 2026.
Paper # MI2025-101 
Date of Issue 2026-02-26 (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 MI2025-101

Conference Information
Committee MI  
Conference Date 2026-03-05 - 2026-03-06 
Place (in Japanese) (See Japanese page) 
Place (in English) OKINAWAKEN SEINENKAIKAN 
Topics (in Japanese) (See Japanese page) 
Topics (in English) Medical Imaging, etc. 
Paper Information
Registration To MI 
Conference Code 2026-03-MI 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) A Study on Masking Ratio and Dataset Expansion for Pre-Training Laparoscopic Image Foundation Models 
Sub Title (in English)  
Keyword(1) Laparoscopic Surgery  
Keyword(2) Vision Transformer  
Keyword(3) Masked Autoencoder  
Keyword(4) Foundation Model  
Keyword(5) Surgical Scene Recognition  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Kaede Yasuda  
1st Author's Affiliation Nagoya University (Nagoya Univ.)
2nd Author's Name Yuichiro Hayashi  
2nd Author's Affiliation Nagoya University (Nagoya Univ.)
3rd Author's Name Masahiro Oda  
3rd Author's Affiliation Nagoya University (Nagoya Univ.)
4th Author's Name Kensaku Mori  
4th Author's Affiliation Nagoya University/National Institute of Informatics (Nagoya Univ./NII/)
5th Author's Name  
5th Author's Affiliation ()
6th Author's Name  
6th Author's Affiliation ()
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 ()
21st Author's Name  
21st Author's Affiliation ()
22nd Author's Name  
22nd Author's Affiliation ()
23rd Author's Name  
23rd Author's Affiliation ()
24th Author's Name  
24th Author's Affiliation ()
25th Author's Name  
25th Author's Affiliation ()
26th Author's Name / /
26th Author's Affiliation ()
()
27th Author's Name / /
27th Author's Affiliation ()
()
28th Author's Name / /
28th Author's Affiliation ()
()
29th Author's Name / /
29th Author's Affiliation ()
()
30th Author's Name / /
30th Author's Affiliation ()
()
31st Author's Name / /
31st Author's Affiliation ()
()
32nd Author's Name / /
32nd Author's Affiliation ()
()
33rd Author's Name / /
33rd Author's Affiliation ()
()
34th Author's Name / /
34th Author's Affiliation ()
()
35th Author's Name / /
35th Author's Affiliation ()
()
36th Author's Name / /
36th Author's Affiliation ()
()
Speaker Author-1 
Date Time 2026-03-06 12:59:00 
Presentation Time 13 minutes 
Registration for MI 
Paper # MI2025-101 
Volume (vol) vol.125 
Number (no) no.395 
Page pp.155-158 
#Pages 4 
Date of Issue 2026-02-26 (MI) 

[Return to Top Page]

[Return to IEICE Web Page]
The Institute of Electronics, Information and Communication Engineers (IEICE), Japan