基于双域N2N和注意力的无监督磁共振成像重建
CSTR:
作者:
作者单位:

(昆明理工大学 信息工程与自动化学院,昆明 650504)

作者简介:

段继忠(1984—),男,副教授,硕士生导师

通讯作者:

段继忠,duanjz@kust.edu.cn

中图分类号:

TP391

基金项目:

国家自然科学基金(61861023);云南省基础研究计划(202301AT070452)


Unsupervised magnetic resonance imaging reconstruction based on dual-domain N2N and attention mechanisms
Author:
Affiliation:

(Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650504, China)

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    为解决目前大部分重建磁共振成像(MRI)的深度学习方法依赖于大量完全采样的数据集进行训练的问题,提出一种包含注意力机制的无监督双域N2N网络(DN2NA)进行并行MRI重建,提出的DN2NA网络不需要额外的训练数据,即可直接对欠采样的k空间数据进行重建。具体而言,在N2N网络中加入复数卷积和通道注意力机制,构建一个基础无监督网络N2NA,同时加入两个物理先验提高频域(k空间)N2NA网络的性能,再将频域N2NA网络与图像域N2NA网络级联成双域网络DN2NA,这种组合方法充分结合了频域和图像域网络的优势。由于在实际应用场景中没有真实数据作为参考,还采用了一种提前停止方法来避免过拟合和提高稳定性。在膝盖和大脑共3个数据集上进行的实验表明,相比现有的无监督网络IUNN和KUNN,DN2NA网络拥有更高的PSNR值和SSIM值以及更低的HFEN值和STD值,这表明DN2NA网络重建质量更好,多次重复重建结果也更稳定。与有监督网络MICCAN相比,DN2NA网络也展现了相似或者更优的性能。

    Abstract:

    To address the reliance of most deep learning methods for magnetic resonance (MR) imaging (MRI) reconstruction on extensive fully-sampled datasets for training, this study proposes an unsupervised dual-domain N2N network with attention mechanisms (DN2NA) for parallel MRI reconstruction. The proposed DN2NA network can directly reconstruct undersampled k-space data without requiring additional training data. Specifically, we integrate complex-valued convolution and channel attention mechanism into the N2N framework to construct a baseline unsupervised network N2NA. Two physical priors are incorporated to enhance the performance of the frequency-domain (k-space) N2NA network, which is then cascaded with an image-domain N2NA network to form the dual-domain DN2NA architecture. This combination effectively leverages the complementary advantages of frequency-domain and image-domain networks. Given the absence of ground-truth references in practical scenarios, an early-stopping strategy is adopted to prevent overfitting and improve stability. Experiments conducted on three knee and brain datasets demonstrate that DN2NA achieves higher PSNR and SSIM, along with lower HFEN and STD compared to existing unsupervised networks (IUNN and KUNN), indicating superior reconstruction quality and stability in repeated reconstructions. Furthermore, DN2NA exhibits comparable or better performance than the supervised network MICCAN.

    参考文献
    相似文献
    引证文献
引用本文

段继忠,陈盛毅.基于双域N2N和注意力的无监督磁共振成像重建[J].哈尔滨工业大学学报,2025,57(10):143. DOI:10.11918/202309074

复制
分享
相关视频

文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2023-09-28
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2025-09-29
  • 出版日期:
文章二维码