ScholarMate
English简体中文繁體中文
客服热线:400-1616-289
登录注册
期刊论文

Structure-preserved meta-learning uniting network for improving low-dose CT quality

Zhu, Manman; Mao, Zerui; Li, Danyang; Wang, Yongbo; Zeng, Dong*; Bian, Zhaoying; Ma, Jianhua*南方医科大学
Physics in Medicine and Biology2022-12-21DOI: 10.1088/1361-6560/aca194
1被引6阅读
引用收藏
分享

摘要

Objective. Deep neural network (DNN) based methods have shown promising performances for low-dose computed tomography (LDCT) imaging. However, most of the DNN-based methods are trained on simulated labeled datasets, and the low-dose simulation algorithms are usually designed based on simple statistical models which deviate from the real clinical scenarios, which could lead to issues of overfitting, instability and poor robustness. To address these issues, in this work, we present a structure-preserved meta-learning uniting network (shorten as 'SMU-Net') to suppress noise-induced artifacts and preserve structure details in the unlabeled LDCT imaging task in real scenarios. Approach. Specifically, the presented SMU-Net contains two networks, i.e., teacher network and student network. The teacher network is trained on simulated labeled dataset and then helps the student network train with the unlabeled LDCT images via the meta-learning strategy. The student network is trained on real LDCT dataset with the pseudo-labels generated by the teacher network. Moreover, the student network adopts the Co-teaching strategy to improve the robustness of the presented SMU-Net. Main results. We validate the proposed SMU-Net method on three public datasets and one real low-dose dataset. The visual image results indicate that the proposed SMU-Net has superior performance on reducing noise-induced artifacts and preserving structure details. And the quantitative results exhibit that the presented SMU-Net method generally obtains the highest signal-to-noise ratio (PSNR), the highest structural similarity index measurement (SSIM), and the lowest root-mean-square error (RMSE) values or the lowest natural image quality evaluator (NIQE) scores. Significance. We propose a meta learning strategy to obtain high-quality CT images in the LDCT imaging task, which is designed to take advantage of unlabeled CT images to promote the reconstruction performance in the LDCT environments.

学科与关键词

关键词
deep neural networklow-dose computed tomographyCo-teaching strategyimage reconstruction
相关
加载中…
  • Deep Neural Networks for Low-dose CT Image Reconstruction via Cooperative Meta-learning Strategy
    Zhu, Manman; Li, Sui; Li, Danyang; Gao, Qi; Zhang, Shanli; Huang, Haiyun; Bian, Zhaoying; Huang, Jing; Zeng, Dong*; Ma, Jianhua*Conference on Medical Imaging - Physics of Medical Imaging, Houston, TX, 2020-02-16 To 2020-02-19.
  • Semi-Supervised Noise Distribution Learning for Low-dose CT Restoration
    Wang, Lei; Gao, Qi; Meng, Mingqiang; Li, Sui; Zhu, Manman; Li, Danyang; Chen, Gaofeng; Zeng, Dong*; Xie, Qi; Zhao, Qian; Bian, Zhaoying; Meng, Deyu; Ma, Jianhua*Conference on Medical Imaging - Physics of Medical Imaging, Houston, TX, 2020-02-16 To 2020-02-19.被引 1
  • Progressive Transfer Learning Strategy for Low-dose CT Image Reconstruction with Limited Annotated Data
    Meng, Mingqiang; Li, Danyang; Li, Sui; Zhu, Manman; Wang, Lei; Gao, Qi; Bian, Zhaoying; Zhang, Xinyu; Huang, Jing; Zeng, Dong*; Ma, Jianhua*Conference on Medical Imaging - Physics of Medical Imaging, Houston, TX, 2020-02-16 To 2020-02-19.
查看全部 →

发表信息

论文状态
公开发表
期刊
Physics in Medicine and Biology
发表日期
2022-12-21
卷期
第 67 卷, 第 24 期
页码
245015
DOI
10.1088/1361-6560/aca194
语言
MULTI-LANGUAGE
期刊等级
2区
成果收录
SCIE
资助标注
NSFC; Science and Technology Program of Guangzhou, China; [U21A6005]; [U1708261]; [81 701 690]; [61 701 217]; [62 001 208]; [202 206 010 148]

作者

  • Z

    Zhu, Manman (第一作者)

    southern med univ

  • M

    Mao, Zerui

    southern med univ

查看全部 7 位作者收起作者
  • L

    Li, Danyang

    southern med univ

  • W

    Wang, Yongbo

    southern med univ

  • Z

    Zeng, Dong (通讯)

    southern med univ

  • B

    Bian, Zhaoying

    southern med univ

  • M

    Ma, Jianhua (通讯)

    southern med univ

登录解锁更多功能

  • 收藏这篇,随时在“我的 · 收藏”找回
  • 认领你的成果,生成推广素材
  • 关注作者获取最新成果
  • 加入项目组、协同写作
注册 / 登录

产品服务

  • 科研之友
  • 创新城
  • 科创云

服务支持

  • 帮助中心
  • 隐私政策
  • 服务条款

联系方式

在线客服:【】
客服热线:400-1616-289
电子邮箱:support@scholarmate.com

关注或下载科研之友

微信二维码
微信公众号
客户端下载二维码
下载客户端
科研成果科研人员科研机构科研动态爱瑞思软件

©2026 深圳市科研之友网络服务有限公司

公安备案图标粤公网安备 44030502000213
粤ICP备 16046710 号粤B2-20110417