Deep Efficient End-to-End Reconstruction (DEER) Network for Few-View Breast CT Image Reconstruction

Breast CT provides image volumes with isotropic resolution in high contrast, enabling detection of small calcification (down to a few hundred microns in size) and subtle density differences. Since breast is sensitive to x-ray radiation, dose reduction of breast CT is an important topic, and for this...

Full description

Bibliographic Details
Main Authors: Huidong Xie, Hongming Shan, Wenxiang Cong, Chi Liu, Xiaohua Zhang, Shaohua Liu, Ruola Ning, Ge Wang
Format: Article
Language:English
Published: IEEE 2020-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/9239986/
_version_ 1818662500075307008
author Huidong Xie
Hongming Shan
Wenxiang Cong
Chi Liu
Xiaohua Zhang
Shaohua Liu
Ruola Ning
Ge Wang
author_facet Huidong Xie
Hongming Shan
Wenxiang Cong
Chi Liu
Xiaohua Zhang
Shaohua Liu
Ruola Ning
Ge Wang
author_sort Huidong Xie
collection DOAJ
description Breast CT provides image volumes with isotropic resolution in high contrast, enabling detection of small calcification (down to a few hundred microns in size) and subtle density differences. Since breast is sensitive to x-ray radiation, dose reduction of breast CT is an important topic, and for this purpose, few-view scanning is a main approach. In this article, we propose a Deep Efficient End-to-end Reconstruction (DEER) network for few-view breast CT image reconstruction. The major merits of our network include high dose efficiency, excellent image quality, and low model complexity. By the design, the proposed network can learn the reconstruction process with as few as (9(N) parameters, where N is the side length of an image to be reconstructed, which represents orders of magnitude improvements relative to the state-of-the-art deeplearning-based reconstruction methods that map raw data to tomographic images directly. Also, validated on a cone-beam breast CT dataset prepared by Koning Corporation on a commercial scanner, our method demonstrates a competitive performance over the state-of-the-art reconstruction networks in terms of image quality. The source code of this paper is available at: https://github.com/HuidongXie/DEER.
first_indexed 2024-12-17T05:01:56Z
format Article
id doaj.art-72daf4fdf183414bacd7ddf16e45f084
institution Directory Open Access Journal
issn 2169-3536
language English
last_indexed 2024-12-17T05:01:56Z
publishDate 2020-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj.art-72daf4fdf183414bacd7ddf16e45f0842022-12-21T22:02:32ZengIEEEIEEE Access2169-35362020-01-01819663319664610.1109/ACCESS.2020.30337959239986Deep Efficient End-to-End Reconstruction (DEER) Network for Few-View Breast CT Image ReconstructionHuidong Xie0https://orcid.org/0000-0002-1124-3548Hongming Shan1https://orcid.org/0000-0002-0604-3197Wenxiang Cong2https://orcid.org/0000-0003-2908-1759Chi Liu3https://orcid.org/0000-0002-7007-1037Xiaohua Zhang4Shaohua Liu5Ruola Ning6Ge Wang7https://orcid.org/0000-0002-2656-7705Department of Biomedical Engineering, Biomedical Imaging Center, Center for Biotechnology and Interdisciplinary Studies, Rensselaer Polytechnic Institute, Troy, NY, USADepartment of Biomedical Engineering, Biomedical Imaging Center, Center for Biotechnology and Interdisciplinary Studies, Rensselaer Polytechnic Institute, Troy, NY, USADepartment of Biomedical Engineering, Biomedical Imaging Center, Center for Biotechnology and Interdisciplinary Studies, Rensselaer Polytechnic Institute, Troy, NY, USADepartment of Biomedical Engineering, Yale University, New Haven, CT, USAKoning Corporation, West Henrietta, NY, USAKoning Corporation, West Henrietta, NY, USAKoning Corporation, West Henrietta, NY, USADepartment of Biomedical Engineering, Biomedical Imaging Center, Center for Biotechnology and Interdisciplinary Studies, Rensselaer Polytechnic Institute, Troy, NY, USABreast CT provides image volumes with isotropic resolution in high contrast, enabling detection of small calcification (down to a few hundred microns in size) and subtle density differences. Since breast is sensitive to x-ray radiation, dose reduction of breast CT is an important topic, and for this purpose, few-view scanning is a main approach. In this article, we propose a Deep Efficient End-to-end Reconstruction (DEER) network for few-view breast CT image reconstruction. The major merits of our network include high dose efficiency, excellent image quality, and low model complexity. By the design, the proposed network can learn the reconstruction process with as few as (9(N) parameters, where N is the side length of an image to be reconstructed, which represents orders of magnitude improvements relative to the state-of-the-art deeplearning-based reconstruction methods that map raw data to tomographic images directly. Also, validated on a cone-beam breast CT dataset prepared by Koning Corporation on a commercial scanner, our method demonstrates a competitive performance over the state-of-the-art reconstruction networks in terms of image quality. The source code of this paper is available at: https://github.com/HuidongXie/DEER.https://ieeexplore.ieee.org/document/9239986/Breast CTdeep learningfew-view CTlow-dose CTX-ray CT
spellingShingle Huidong Xie
Hongming Shan
Wenxiang Cong
Chi Liu
Xiaohua Zhang
Shaohua Liu
Ruola Ning
Ge Wang
Deep Efficient End-to-End Reconstruction (DEER) Network for Few-View Breast CT Image Reconstruction
IEEE Access
Breast CT
deep learning
few-view CT
low-dose CT
X-ray CT
title Deep Efficient End-to-End Reconstruction (DEER) Network for Few-View Breast CT Image Reconstruction
title_full Deep Efficient End-to-End Reconstruction (DEER) Network for Few-View Breast CT Image Reconstruction
title_fullStr Deep Efficient End-to-End Reconstruction (DEER) Network for Few-View Breast CT Image Reconstruction
title_full_unstemmed Deep Efficient End-to-End Reconstruction (DEER) Network for Few-View Breast CT Image Reconstruction
title_short Deep Efficient End-to-End Reconstruction (DEER) Network for Few-View Breast CT Image Reconstruction
title_sort deep efficient end to end reconstruction deer network for few view breast ct image reconstruction
topic Breast CT
deep learning
few-view CT
low-dose CT
X-ray CT
url https://ieeexplore.ieee.org/document/9239986/
work_keys_str_mv AT huidongxie deepefficientendtoendreconstructiondeernetworkforfewviewbreastctimagereconstruction
AT hongmingshan deepefficientendtoendreconstructiondeernetworkforfewviewbreastctimagereconstruction
AT wenxiangcong deepefficientendtoendreconstructiondeernetworkforfewviewbreastctimagereconstruction
AT chiliu deepefficientendtoendreconstructiondeernetworkforfewviewbreastctimagereconstruction
AT xiaohuazhang deepefficientendtoendreconstructiondeernetworkforfewviewbreastctimagereconstruction
AT shaohualiu deepefficientendtoendreconstructiondeernetworkforfewviewbreastctimagereconstruction
AT ruolaning deepefficientendtoendreconstructiondeernetworkforfewviewbreastctimagereconstruction
AT gewang deepefficientendtoendreconstructiondeernetworkforfewviewbreastctimagereconstruction