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

DeTransUnet: attenuation correction of gated cardiac images without structural information

Wang, Bo; Lu, Lijun*; Liu, Huafeng*南方医科大学; 浙江大学
Physics in Medicine and Biology2022-8-21DOI: 10.1088/1361-6560/ac840e
2被引8阅读
引用收藏
分享

摘要

Objective. Myocardial perfusion imaging (MPI) with positron emission tomography (PET) is a non-invasive imaging method, and it is of great significance to the diagnosis and prognosis of coronary heart disease. Attenuation correction (AC) for PET images is a necessary step for further quantitative analysis. In order not to use magnetic resonance (MR) or computed tomography (CT) images for AC, this work proposes DeTransUnet to obtain AC PET images directly from no-attenuation corrected (NAC) PET images. Approach. The proposed DeTransUnet is a 3D structure which combines the multi-scale deformable transformer layers and the 3D convolutional neural network (CNN). And it integrates the advantages of transformer with long-range dependence and CNN suitable for image calculation. The AC images using CT images for AC and scatter correction (SC) and are considered as training labels, while the NAC images are reconstructed without AC and SC. Standard uptake value (SUV) values are calculated for both NAC and AC images to exclude the influence of weight and injection dose. With NAC SUV images as the inputs of the DeTransUnet, the outputs of DeTransUnet are AC SUV images. Main results. The proposed DeTransUnet was performed on an MPI gated-PET dataset, and the results were compared with Unet2D and Unet2.5D. The metrics of the whole image and the left ventricular myocardium show that the proposed method has advantages over other deep learning methods. Significance. The proposed DeTransUnet is a novel AC framework that does not require CT or MR images. It can be used as an independent AC method on PET/MR instrument. In addition, when CT images contain defects or cannot be registered with PET images on PET/CT instrument, DeTransUnet is able to repair the defects and keep consistent with the NAC images.

学科与关键词

关键词
deep learningPETattenuation correctioncoronary heart diseasemyocardial perfusion imaging
相关
加载中…
  • INVERTIBLE AC-FLOW: DIRECT ATTENUATION CORRECTION OF PET IMAGES WITHOUT CT OR MR IMAGES
    Wang, Bo; Lu, Lijun*; Liu, Huafeng*19th IEEE International Symposium on Biomedical Imaging (IEEE ISBI), Kolkata, INDIA, 2022-03-28 To 2022-03-31.被引 1
  • Deep-learning-based methods of attenuation correction for SPECT and PET
    Chen, Xiongchao; Liu, Chi*Journal of Nuclear Cardiology, 2023, 30(5): 1859-1878.被引 13
  • Transfer learning-based attenuation correction for static and dynamic cardiac PET using a generative adversarial network
    Sun, Hao; Wang, Fanghu; Yang, Yuling; Hong, Xiaotong; Xu, Weiping; Wang, Shuxia*; Mok, Greta S. P.*; Lu, Lijun*European Journal of Nuclear Medicine and Molecular Imaging, 2023, 50(12): 3630-3646.被引 1
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
2022-08-21
卷期
第 67 卷, 第 16 期
页码
165007
DOI
10.1088/1361-6560/ac840e
语言
MULTI-LANGUAGE
期刊等级
2区
成果收录
SCIE
资助标注
National Natural Science Foundation of China [U1809204, 61525106, 81871437]; National Key Technology Research and Development Program of China [2017YFE0104000]; Key Research and Development Program of Zhejiang Province [2021C03029]; Talent Program of Zhejiang Province [2021R51004]

作者

  • W

    Wang, Bo (第一作者)

    zhejiang univ

  • L

    Lu, Lijun (通讯)

    southern med univ

查看全部 3 位作者收起作者
  • L

    Liu, Huafeng (通讯)

    zhejiang 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