Presentation | 2019-01-23 Influence of group normalization in multi-class organ segmentation of abdominal CT volumes Chen Shen, Fausto Milletari, Holger R. Roth, Hirohisa Oda, Masahiro Oda, Yuichiro Hayashi, Kazunari Misawa, Kensaku Mori, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Organ segmentation is one of the most important branches of medical image analysis. Fully convolutional networks (FCNs) have become the dominant approach for this task and achieved considerable improvements for automated organ segmentation in volumetric image data, such as computed tomography images. In this paper, we investigate the influence of group normalization (GN) in multi-class organ segmentation from 3D CT volumes using fully convolutional network. Batch normalization is widely utilized in deep learning based methods to accelerate the convergence, reduce the reliance on initial learning rate and avoid overfitting. However, this type of normalization is strongly related to the batch size. Here, we study the influence of GN which is independent from batch size. In this research, we performed experiments on 377 cases of portal vein-phase abdominal CT volumes. The segmentation performance for small organs like artery and pancreas improved by introducing GN. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | deep learning / multi-organ segmentation / group normalization / computed tomography |
Paper # | MI2018-94 |
Date of Issue | 2019-01-15 (MI) |
Conference Information | |
Committee | MI |
---|---|
Conference Date | 2019/1/22(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Medical Image Engineering, Analysis, Recognition, etc. |
Chair | Kensaku Mori(Nagoya Univ.) |
Vice Chair | Yoshiki Kawata(Tokushima Univ.) / Yuichi Kimura(Kinki Univ.) |
Secretary | Yoshiki Kawata(Aichi Inst. of Tech.) / Yuichi Kimura(Nagoya Inst. of Tech.) |
Assistant | Ryo Haraguchi(Univ. of Hyogo) / Yasushi Hirano(Yamaguchi Univ.) |
Paper Information | |
Registration To | Medical Imaging |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Influence of group normalization in multi-class organ segmentation of abdominal CT volumes |
Sub Title (in English) | |
Keyword(1) | deep learning |
Keyword(2) | multi-organ segmentation |
Keyword(3) | group normalization |
Keyword(4) | computed tomography |
1st Author's Name | Chen Shen |
1st Author's Affiliation | Nagoya University(Nagoya Univ.) |
2nd Author's Name | Fausto Milletari |
2nd Author's Affiliation | Nvidia(Nvidia) |
3rd Author's Name | Holger R. Roth |
3rd Author's Affiliation | Nvidia(Nvidia) |
4th Author's Name | Hirohisa Oda |
4th Author's Affiliation | Nagoya University(Nagoya Univ.) |
5th Author's Name | Masahiro Oda |
5th Author's Affiliation | Nagoya University(Nagoya Univ.) |
6th Author's Name | Yuichiro Hayashi |
6th Author's Affiliation | Nagoya University(Nagoya Univ.) |
7th Author's Name | Kazunari Misawa |
7th Author's Affiliation | Aichi Cancer Center Hospital(Aichi Cancer Center Hospital) |
8th Author's Name | Kensaku Mori |
8th Author's Affiliation | Nagoya University(Nagoya Univ.) |
Date | 2019-01-23 |
Paper # | MI2018-94 |
Volume (vol) | vol.118 |
Number (no) | MI-412 |
Page | pp.pp.143-148(MI), |
#Pages | 6 |
Date of Issue | 2019-01-15 (MI) |