• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

A study of sparse representation based computed tomography

Research Project

Project/Area Number 16K00328
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Research Field Soft computing
Research InstitutionThe University of Electro-Communications

Principal Investigator

Shouno Hayaru  電気通信大学, 大学院情報理工学研究科, 教授 (50263231)

Co-Investigator(Kenkyū-buntansha) 坂田 宗之  地方独立行政法人東京都健康長寿医療センター(東京都健康長寿医療センター研究所), 東京都健康長寿医療センター研究所, 研究員 (00403329)
Project Period (FY) 2016-04-01 – 2019-03-31
Project Status Completed (Fiscal Year 2018)
Budget Amount *help
¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2018: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2017: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
Fiscal Year 2016: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
Keywordsスパースモデリング / 断層画像再構成 / ノイズ除去 / 辞書学習 / PET / ラドン変換 / ノイズ低減 / PET画像 / スパース表現 / ディープラーニング / PET画像再構成 / スパース基底 / 深層学習 / 医用画像再構成 / Deep Learning / 生成モデル
Outline of Final Research Achievements

We applied sparse modeling techniques for the image reconstruction algorithm for Computed tomography (CT) image, such like as Positron Emission Tomography(PET) or X-ray CT image. We applied two types of sparse representations. One is for image domain and the other is for observation domain called Radon space. In image domain, we adopt total variation (TV) method, and K-SVD (singular value decomposition) method, which is in a category of dictionary learning. In order to construct a dictionary component, we apply Radon transformed natural images. By use of two types of sparse representations, we confirm the reconstruction performance is improved under the noise environment.

Academic Significance and Societal Importance of the Research Achievements

X線CT画像やPET画像は,非破壊検査の一環として用いられるが,十分な強さの線源を用いなければ,再構成画像のアーティファクトと呼ばれるノイズがのる.その一方で,あまり強い線源を用いた場合,観測対象の被曝量が問題になるケースがある.そこで低被爆でよりよい再構成画像を得るためのノイズ除去をスパース表現を用いて構築した.この意義は検査を行う対象をより安全にするための手法として有効であると考えている.

Report

(4 results)
  • 2018 Annual Research Report   Final Research Report ( PDF )
  • 2017 Research-status Report
  • 2016 Research-status Report
  • Research Products

    (16 results)

All 2019 2018 2017 2016

All Journal Article (7 results) (of which Int'l Joint Research: 1 results,  Peer Reviewed: 4 results,  Acknowledgement Compliant: 1 results) Presentation (9 results) (of which Int'l Joint Research: 6 results,  Invited: 2 results)

  • [Journal Article] ベイズ的変数選択法に基づく分光スペクトル分解2019

    • Author(s)
      川島 貴大,庄野 逸
    • Journal Title

      情報処理学会論文誌「数理モデル化と応用」

      Volume: 印刷中

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] 2)Feature Representation Analysis of Deep Convolutional Neural Network using Two-stage Feature Transfer―An Application for Diffuse Lung Disease Classification2018

    • Author(s)
      Aiga Suzuki, Hidenori Sakanashi, Shoji Kido, Hayaru Shouno
    • Journal Title

      情報処理学会論文誌「数理モデル化と応用」

      Volume: 11 Pages: 74-83

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] 1)Simultaneous Estimation of Nongaussian Components and Their Correlation Structure2017

    • Author(s)
      Hiroaki Sasaki, Michael U. Gutmann, Hayaru Shouno, Aapo Hyvarinen
    • Journal Title

      Neural Computation

      Volume: 29 Issue: 11 Pages: 1-38

    • DOI

      10.1162/neco_a_01006

    • Related Report
      2017 Research-status Report
    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] 2)Support Vector Machine Histogram: New Analysis and Architecture Design Method of Deep Convolutional Neural Network2017

    • Author(s)
      Satoshi Suzuki, Hayaru Shouno
    • Journal Title

      Neural Processing Letters

      Volume: 1 Issue: 3 Pages: 1-16

    • DOI

      10.1007/s11063-017-9652-0

    • Related Report
      2017 Research-status Report
    • Peer Reviewed
  • [Journal Article] 1)脳情報科学が拓くAIとICT:2.脳情報科学と人工知能 -ネオコグニトロンからDeep Learningへ-2017

    • Author(s)
      本武陽一,庄野逸,田村弘,岡田真人
    • Journal Title

      情報処理

      Volume: 59 Pages: 42-47

    • Related Report
      2017 Research-status Report
  • [Journal Article] ディープラーニングの概要と医療分野への応用2017

    • Author(s)
      庄野 逸
    • Journal Title

      インナービジョン

      Volume: 32 Pages: 7-9

    • Related Report
      2017 Research-status Report
  • [Journal Article] スパースモデリングの歴史と基本技術2016

    • Author(s)
      庄野逸
    • Journal Title

      電子情報通信学会論文誌

      Volume: 99(5) Pages: 376-380

    • Related Report
      2016 Research-status Report
    • Acknowledgement Compliant
  • [Presentation] Mosquito Larva Classification based on a Convolution Neural Network2018

    • Author(s)
      2)Alejandra Sanchez, Mariko Nakano, Henrik Tunnermann, Toya Teramoto, Hayaru Shouno
    • Organizer
      PDPTA2018
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research
  • [Presentation] 辞書学習を用いたPET画像再構成2018

    • Author(s)
      奥村 直裕,庄野 逸
    • Organizer
      電子情報通信学会ニューロコンピューティング研究会
    • Related Report
      2018 Annual Research Report
  • [Presentation] Generative Model of Textures using Hierarchical Probabilistic Principal Components2017

    • Author(s)
      Aiga Suzuki, Hayaru Shouno
    • Organizer
      International Conference on Parallel Distributed Processing Techniques and Applications
    • Related Report
      2017 Research-status Report
    • Int'l Joint Research
  • [Presentation] Comparison of Feature Selection Method for Diffuse Lung Disease2017

    • Author(s)
      Satoshi Ono, Makoto Koiwai, Hayaru Shouno, Shoji Kido
    • Organizer
      International Conference on Parallel Distributed Processing Techniques and Applications
    • Related Report
      2017 Research-status Report
    • Int'l Joint Research
  • [Presentation] A study on Visual Interpretation of Network in Network2017

    • Author(s)
      Satoshi Suzuki, Hayaru Shouno
    • Organizer
      International Joint Conference on Neural Networks
    • Related Report
      2017 Research-status Report
    • Int'l Joint Research
  • [Presentation] 医用画像識別におけるスパース特徴選択手法について2017

    • Author(s)
      庄野 逸
    • Organizer
      電子情報通信学会ソサエティ大会
    • Related Report
      2017 Research-status Report
    • Invited
  • [Presentation] Medical Texture Image Classification using Deep Convolution Neural Network with Transfer Style learning2017

    • Author(s)
      庄野 逸
    • Organizer
      第27回日本神経回路学会シンポジウム「スパースモデリングの深化と高次元データ駆動科学の創成」
    • Related Report
      2017 Research-status Report
    • Invited
  • [Presentation] Feature Selection for Diffuse Lung Disease using Exchange Markov Chain Monte-Carlo Method2016

    • Author(s)
      Makoto Koiwai, Nodoka Iida, Hayaru Shouno, Shoji Kido
    • Organizer
      PDPTA2016
    • Place of Presentation
      Las Vegas (USA)
    • Year and Date
      2016-07-25
    • Related Report
      2016 Research-status Report
    • Int'l Joint Research
  • [Presentation] Architecture Design of Deep Convolutional Neural Network for Diffuse Lung Disease Using Representation Separation Information2016

    • Author(s)
      Satoshi Suzuki, Nodoka Iida, Hayaru Shouno, Shoji Kido
    • Organizer
      PDPTA2016
    • Place of Presentation
      Las Vegas (USA)
    • Year and Date
      2016-07-25
    • Related Report
      2016 Research-status Report
    • Int'l Joint Research

URL: 

Published: 2016-04-21   Modified: 2020-03-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi