Multiclass Dictionary-Based Statistical Iterative Reconstruction for Low-Dose CT

Hiryu KAMOSHITA; Daichi KITAHARA; Ken'ichi FUJIMOTO; Laurent CONDAT; Akira HIRABAYASHI

doi:10.1587/transfun.2020EAP1020

Multiclass Dictionary-Based Statistical Iterative Reconstruction for Low-Dose CT

Hiryu KAMOSHITA, Daichi KITAHARA, Ken'ichi FUJIMOTO, Laurent CONDAT, Akira HIRABAYASHI

Full Text Views

0

Cite this

Summary :

This paper proposes a high-quality computed tomography (CT) image reconstruction method from low-dose X-ray projection data. A state-of-the-art method, proposed by Xu et al., exploits dictionary learning for image patches. This method generates an overcomplete dictionary from patches of standard-dose CT images and reconstructs low-dose CT images by minimizing the sum of a data fidelity and a regularization term based on sparse representations with the dictionary. However, this method does not take characteristics of each patch, such as textures or edges, into account. In this paper, we propose to classify all patches into several classes and utilize an individual dictionary with an individual regularization parameter for each class. Furthermore, for fast computation, we introduce the orthogonality to column vectors of each dictionary. Since similar patches are collected in the same cluster, accuracy degradation by the orthogonality hardly occurs. Our simulations show that the proposed method outperforms the state-of-the-art in terms of both accuracy and speed.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E104-A No.4 pp.702-713

Publication Date: 2021/04/01

Publicized: 2020/10/06

Online ISSN: 1745-1337

DOI: 10.1587/transfun.2020EAP1020

Type of Manuscript: PAPER

Category: Numerical Analysis and Optimization

Authors

Hiryu KAMOSHITA
  Ritsumeikan University
Daichi KITAHARA
  Ritsumeikan University
Ken'ichi FUJIMOTO
  Kagawa University
Laurent CONDAT
  King Abdullah University of Science and Technology
Akira HIRABAYASHI
  Ritsumeikan University

Keyword

low-dose CT, image reconstruction, sparse representation, fast dictionary learning, clustering

Cite this

Copy

Hiryu KAMOSHITA, Daichi KITAHARA, Ken'ichi FUJIMOTO, Laurent CONDAT, Akira HIRABAYASHI, "Multiclass Dictionary-Based Statistical Iterative Reconstruction for Low-Dose CT" in IEICE TRANSACTIONS on Fundamentals, vol. E104-A, no. 4, pp. 702-713, April 2021, doi: 10.1587/transfun.2020EAP1020.
Abstract: This paper proposes a high-quality computed tomography (CT) image reconstruction method from low-dose X-ray projection data. A state-of-the-art method, proposed by Xu et al., exploits dictionary learning for image patches. This method generates an overcomplete dictionary from patches of standard-dose CT images and reconstructs low-dose CT images by minimizing the sum of a data fidelity and a regularization term based on sparse representations with the dictionary. However, this method does not take characteristics of each patch, such as textures or edges, into account. In this paper, we propose to classify all patches into several classes and utilize an individual dictionary with an individual regularization parameter for each class. Furthermore, for fast computation, we introduce the orthogonality to column vectors of each dictionary. Since similar patches are collected in the same cluster, accuracy degradation by the orthogonality hardly occurs. Our simulations show that the proposed method outperforms the state-of-the-art in terms of both accuracy and speed.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.2020EAP1020/_p

Copy

@ARTICLE{e104-a_4_702,
author={Hiryu KAMOSHITA, Daichi KITAHARA, Ken'ichi FUJIMOTO, Laurent CONDAT, Akira HIRABAYASHI, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Multiclass Dictionary-Based Statistical Iterative Reconstruction for Low-Dose CT},
year={2021},
volume={E104-A},
number={4},
pages={702-713},
abstract={This paper proposes a high-quality computed tomography (CT) image reconstruction method from low-dose X-ray projection data. A state-of-the-art method, proposed by Xu et al., exploits dictionary learning for image patches. This method generates an overcomplete dictionary from patches of standard-dose CT images and reconstructs low-dose CT images by minimizing the sum of a data fidelity and a regularization term based on sparse representations with the dictionary. However, this method does not take characteristics of each patch, such as textures or edges, into account. In this paper, we propose to classify all patches into several classes and utilize an individual dictionary with an individual regularization parameter for each class. Furthermore, for fast computation, we introduce the orthogonality to column vectors of each dictionary. Since similar patches are collected in the same cluster, accuracy degradation by the orthogonality hardly occurs. Our simulations show that the proposed method outperforms the state-of-the-art in terms of both accuracy and speed.},
keywords={},
doi={10.1587/transfun.2020EAP1020},
ISSN={1745-1337},
month={April},}

Copy

TY - JOUR
TI - Multiclass Dictionary-Based Statistical Iterative Reconstruction for Low-Dose CT
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 702
EP - 713
AU - Hiryu KAMOSHITA
AU - Daichi KITAHARA
AU - Ken'ichi FUJIMOTO
AU - Laurent CONDAT
AU - Akira HIRABAYASHI
PY - 2021
DO - 10.1587/transfun.2020EAP1020
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E104-A
IS - 4
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - April 2021
AB - This paper proposes a high-quality computed tomography (CT) image reconstruction method from low-dose X-ray projection data. A state-of-the-art method, proposed by Xu et al., exploits dictionary learning for image patches. This method generates an overcomplete dictionary from patches of standard-dose CT images and reconstructs low-dose CT images by minimizing the sum of a data fidelity and a regularization term based on sparse representations with the dictionary. However, this method does not take characteristics of each patch, such as textures or edges, into account. In this paper, we propose to classify all patches into several classes and utilize an individual dictionary with an individual regularization parameter for each class. Furthermore, for fast computation, we introduce the orthogonality to column vectors of each dictionary. Since similar patches are collected in the same cluster, accuracy degradation by the orthogonality hardly occurs. Our simulations show that the proposed method outperforms the state-of-the-art in terms of both accuracy and speed.
ER -