Presentation 2019-10-19
Localization of Diffuse Lung Deseases' Lesions and Quantification of Their Volumes Using Deep Learning
Hiroaki Takebe, Yasutaka Moriwaki, Nobuhiro Miyazaki, Takayuki Baba, Hiroaki Terada, Toru Higaki, Kazuo Awai, Hirotaka Kobayashi, Machiko Nakagawa, Masahiko Shimada, Kenji Kitayama,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Changes in the amount of lesion over time are important information for diagnostic imaging of diffuse lung disease in which lesions appear throughout a lung. Then a technique that can automatically quantify the amount of lesion is desired. The quantification of the amount of lesion is performed by classifying the lesion from the image. A method using deep learning (DL) is effective for classifying lesions in diffuse lung disease. However, there is a problem that it is difficult to classify micro nodules that are typical lesions. We propose a method for accurately classifying micro nodules using DL by focusing on the 3D structure of a micro nodule. We confirmed that as for the proposed method, the classification accuracy was improved compared to the conventional method, and as an application of this method, the temporal change in the amount of lesions was accurately quantified for cases taken at multiple time points of the same patient.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Deep Learning / Diffuse Lung Disease / Lesion Classification / Micro Nodule
Paper # PRMU2019-44
Date of Issue 2019-10-11 (PRMU)

Conference Information
Committee PRMU
Conference Date 2019/10/18(2days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Yoichi Sato(Univ. of Tokyo)
Vice Chair Toru Tamaki(Hiroshima Univ.) / Akisato Kimura(NTT)
Secretary Toru Tamaki(NTT) / Akisato Kimura(OMRON SINICX)
Assistant Yusuke Uchida(DeNA) / Takayoshi Yamashita(Chubu Univ.)

Paper Information
Registration To Technical Committee on Pattern Recognition and Media Understanding
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Localization of Diffuse Lung Deseases' Lesions and Quantification of Their Volumes Using Deep Learning
Sub Title (in English)
Keyword(1) Deep Learning
Keyword(2) Diffuse Lung Disease
Keyword(3) Lesion Classification
Keyword(4) Micro Nodule
1st Author's Name Hiroaki Takebe
1st Author's Affiliation FUJITSU LABORATORIES LTD.(FUJITSU LAB.)
2nd Author's Name Yasutaka Moriwaki
2nd Author's Affiliation FUJITSU LABORATORIES LTD.(FUJITSU LAB.)
3rd Author's Name Nobuhiro Miyazaki
3rd Author's Affiliation FUJITSU LABORATORIES LTD.(FUJITSU LAB.)
4th Author's Name Takayuki Baba
4th Author's Affiliation FUJITSU LABORATORIES LTD.(FUJITSU LAB.)
5th Author's Name Hiroaki Terada
5th Author's Affiliation Hiroshima University(Hiroshima Univ.)
6th Author's Name Toru Higaki
6th Author's Affiliation Hiroshima University(Hiroshima Univ.)
7th Author's Name Kazuo Awai
7th Author's Affiliation Hiroshima University(Hiroshima Univ.)
8th Author's Name Hirotaka Kobayashi
8th Author's Affiliation FUJITSU LIMITED(FUJITSU)
9th Author's Name Machiko Nakagawa
9th Author's Affiliation FUJITSU LIMITED(FUJITSU)
10th Author's Name Masahiko Shimada
10th Author's Affiliation FUJITSU LIMITED(FUJITSU)
11th Author's Name Kenji Kitayama
11th Author's Affiliation FUJITSU LIMITED(FUJITSU)
Date 2019-10-19
Paper # PRMU2019-44
Volume (vol) vol.119
Number (no) PRMU-235
Page pp.pp.67-72(PRMU),
#Pages 6
Date of Issue 2019-10-11 (PRMU)