Presentation 2018-03-19
A preliminary study on unsupervised registration with deep learning
Kai Nagara, Holger R. Roth, Shota Nakamura, Masahiro Oda, Kensaku Mori,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Registration is one of the important processes in medical image processing. Many researchers have proposed methods for registration. However, few learning-based non-rigid registration method has developed because of the difficulty of preparing a lot of ground truth data for non-rigid registration. To solve this problem, a registration method by using unsupervised deep learning has been proposed. In the registration method by unsupervised learning, it is not necessary to create ground truth data. However, in this method, the network cannot learn the global structures because the range of convolution is limited. Therefore, in this paper, we propose a network introducing a coarse-to-fine approach to learn global structures. The proposed network consists of multiple networks with pooling layers having different kernel sizes. Moreover, we develop a loss function that combines outputs from these multiple networks. The proposed network outputs deformed images according to the reference images. Experimental results using the proposed method showed better registration results than conventional methods.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Deep learning / Unsupervised learning / Non-rigid registration / Micro CT / HE stained image
Paper # MI2017-65
Date of Issue 2018-03-12 (MI)

Conference Information
Committee SIP / EA / SP / MI
Conference Date 2018/3/19(2days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English) Speech, Engineering/Electro Acoustics, Signal Processing, and Related Topics [SIP, EA, SP]/ Medical Image Engineering, Analysis, Recognition, etc. [MI]
Chair Masahiro Okuda(Univ. of Kitakyushu) / Suehiro Shimauchi(NTT) / Yoichi Yamashita(Ritsumeikan Univ.) / Kensaku Mori(Nagoya Univ.)
Vice Chair Shogo Muramatsu(Niigata Univ.) / Naoyuki Aikawa(TUS) / Mitsunori Mizumachi(Kyutech) / Hiroki Mori(Utsunomiya Univ.) / Yoshiki Kawata(Tokushima Univ.) / Yuichi Kimura(Kinki Univ.)
Secretary Shogo Muramatsu(Chiba Inst. of Tech.) / Naoyuki Aikawa(Takushoku Univ.) / Mitsunori Mizumachi(Akita Pref. Univ.) / Hiroki Mori(Shizuoka Inst. of Science and Tech.) / Yoshiki Kawata(Shizuoka Univ.) / Yuichi Kimura(Meijo Univ.)
Assistant Masayoshi Nakamoto(Hiroshima Univ.ひろ) / TREVINO Jorge(Tohoku Univ.) / Nobutaka Ito(NTT) / Kei Hashimoto(Nagoya Inst. of Tech.) / Satoshi Kobashikawa(NTT) / Ryo Haraguchi(Univ. of Hyogo) / Yasushi Hirano(Yamaguchi Univ.)

Paper Information
Registration To Technical Committee on Signal Processing / Technical Committee on Engineering Acoustics / Technical Committee on Speech / Technical Committee on Medical Imaging
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A preliminary study on unsupervised registration with deep learning
Sub Title (in English)
Keyword(1) Deep learning
Keyword(2) Unsupervised learning
Keyword(3) Non-rigid registration
Keyword(4) Micro CT
Keyword(5) HE stained image
1st Author's Name Kai Nagara
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 Shota Nakamura
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 Kensaku Mori
5th Author's Affiliation Nagoya University(Nagoya Univ.)
Date 2018-03-19
Paper # MI2017-65
Volume (vol) vol.117
Number (no) MI-518
Page pp.pp.7-12(MI),
#Pages 6
Date of Issue 2018-03-12 (MI)