[image 00136] IEEE-EMBC：医用画像チュートリアル＆ワークショップ（7/3大阪国際会議場）

[Yoshinobu Sato]

メーリングリストの皆様
（重複の場合は，ご容赦ください．）

大阪大学の佐藤嘉伸です．

IEEE EBMC 国際会議 ( http://embc2013.embs.org/ ) の一環として，
7月3日（水）大阪国際会議場にて，
　チュートリアル「医用画像における統計的モデリングと機械学習」を午前，
　ワークショップ「全身計算解剖学とその診断治療支援応用」を午後，
同じ会場にて開催いたします．

医用画像分野における最先端の技術開発と研究成果を1日で見ることが
できる貴重な機会です．皆様奮ってご参加下さい．

以下で参加登録をしていただけます．会議本体に参加されない場合でも，
チュートリアル・ワークショップのみに参加登録することができます．
http://embc2013.embs.org/registration.html

登録料については以下をご参照ください．
http://embc2013.embs.org/wstut.html


詳細につきましては、以下をご参照ください．
--------------------------------------------------------------
Workshops & Tutorials ( http://embc2013.embs.org/wstut.html )
35th Annual International Conference of IEEE Engineering in Medicine and
Biology Society (IEEE EMBC 2013)
( http://embc2013.embs.org/index.html )
Place: Osaka International Convention Center, Osaka, Japan
Date: July 3, 2013


Tutorial title: Statistical Modeling and Machine Learning in Medical Imaging
( http://embc2013.embs.org/pdf/tut6.pdf )
Time: 8:30am-12:30am

Organizer: Kenji Suzuki (University of Chicago, USA)

Tutorial topics and speakers:
- “Fundamentals and Applications of Super Resolution in Medical Imaging”
by Yen-Wei Chen (Ritsumeikan University, Japan)
- “Fundamentals of Statistical Image Reconstruction in X-ray CT, SPECT
and PET” by Hiroyuki Kudo (University of Tsukuba, Japan)
- “Pixel-based Machine Learning (PML) in Medical Image Processing and
Computer-aided Diagnosis” by Kenji Suzuki (University of Chicago, USA)

Abstract:
Statistical modeling and machine learning play important roles in the
medical imaging field, including image reconstruction, image processing
and analysis, and computer-aided diagnosis and therapy. As medical
imaging modalities are advanced, the amount of data from modalities
increases dramatically. Such modalities include cone-beam CT, 3D
ultrasound imaging, diffusion/functional/interventional MRI, and
PET-CT/MRI. Consequently, algorithms/methods which can handle such a
large amount of data are demanded. Furthermore, it is difficult to
derive analytic solutions to represent lesions/anatomy because of large
variations and complexity. Therefore, tasks in medical imaging require
statistical representation of and/or learning from data.

Thus, statistical modeling and machine learning obtain enthusiastic
attentions from the community. Statistical modeling is a modeling
technique for representing data based on statistical methods such as
Bayesian methods, Gibbs sampling, and principal/independent component
analysis. On the other hand, machine learning is a technique aiming at
acquiring functions/knowledge/tasks/models through “learning from
examples/data,” including neural networks, support vector machines,
manifold learning, and dictionary learning.

This tutorial will provide the overview, fundamentals, and applications
of statistical modeling and machine learning to three important areas:
image reconstruction, super-resolution, and computer-aided diagnosis. We
invite 3 experts in the areas as instructors in this tutorial. Dr. Kudo
will talk about iterative (or statistical) reconstruction that is a
mainstream research in the area, its basic principles, and advantages
over analytical reconstruction. Dr. Chen will talk about
super-resolution techniques that convert low-resolution images to
high-resolution images by using machine learning, its theory, recent
advances, and applications. Dr. Suzuki will talk about pixel-based
machine learning that learns pixels/images directly, as opposed to
extracted features from segmented lesions, its fundamentals, advances,
and applications in computer-aided diagnosis.
Thus, the tutorials provide excellent opportunities for attendees to
learn from the fundamentals to specific applications in the field of
statistical modeling and machine learning in medical imaging.


Workshop title: Whole-Body Computational Anatomy and its Application to
Computer Aided Diagnosis and Therapy
( http://embc2013.embs.org/pdf/ws10.pdf )
Time: 13:30pm-17:30pm

Organizers: Yoshinobu Sato, Makoto Hashizume

Topics and speakers
Whole-body Computational Anatomy: Mathematical Foundations and Basic
Technologies
- "Statistical landmark modeling and detection" by Y. Masutani (U Tokyo)
- "Fast organ localization and FDG-PET modeling" by H. Fujita (Gifu Univ)
- "Multi-organ modeling and segmentation" by Y. Sato (Osaka Univ)
- "Multi-scale organ modeling" by N. Niki (U Tokushima)
Computational Anatomy for Diagnostic and Therapeutic Assistance
- "Technical developments" by K. Mori (Nagoya Univ)
- "Clinical perspectives" by M. Hashizume (Kyushu Univ)
Computational Anatomy for Autopsy Imaging
- "Technical developments" by A. Shimizu (TUAT)
- "Clinical perspectives" by S. Kido (Yamaguchi Univ)

Abstract
Computational anatomy provides effective means to better understand
anatomical variability, to support the diagnosis of disease, to simulate
realistically intervention and so on. The depth and spectrum of
technological topics in computational anatomy have expanded to encompass
all aspects of intelligent segmentation and modeling of complex objects,
understanding of 3D images, advanced pattern recognition, man-machine
interface technologies, virtual reality technologies, and so on.
However, its target has been focused primarily on the brain.
Technologies of computational anatomy for the whole body including the
chest and abdomen are still at a lower level compared with that for the
brain. To close this gap, a research project on “Computational Anatomy
for Computer-Aided Diagnosis and Therapy: Frontiers of Medical Image
Sciences (Computational Anatomy in short)” has been organized in 2009.
It is supported by the Grant-in-Aid for Scientific Research on
Innovative Areas from the Ministry of Education, Culture, Sports,
Science and Technology, Japan. This project aims to establish a new
discipline which provides a mathematical framework to deal with human
anatomy primarily focused on chest, abdomen, and musculoskeletal system
based on medical images. The challenges consist of (1) development of
theories for representation of anatomical models that cover
inter-individual variability in shape and topology and its construction
through statistical analysis of population data, (2) investigation of
methodologies for precise and robust retrieval of anatomical information
from medical images, virtually equivalent to real human body dissection,
and (3) development of innovative technologies assisting medical
diagnosis and interventions based on computational anatomy. The outcomes
are expected to contribute to advanced medicine, basic biomedical
research, medical education, and information science. In this workshop,
the progress of the “Computational Anatomy” project is overviewed.
--------------------------------------------------------------

-- 
Yoshinobu Sato, Ph.D
Associate Professor
Image Analysis Group
Department of Diagnostic and Interventional Radiology
Osaka University Graduate School of Medicine
Tel: +81(0)6-6879-3562, Fax: +81(0)6-6879-3569
yoshi ＠ image.med.osaka-u.ac.jp
yoshinobu.sato ＠ ieee.org
http://www.image.med.osaka-u.ac.jp/yoshi/