[1]刘环宇,郭海鹏,刘晓东,等.基于隐式退化模型的磁共振图像超分辨重建网络[J].中国医学物理学杂志,2024,41(6):690-701.[doi:DOI:10.3969/j.issn.1005-202X.2024.06.006]
 LIU Huanyu,GUO Haipeng,LIU Xiaodong,et al.Super-resolution reconstruction network based on implicit degradation model for magnetic resonance images[J].Chinese Journal of Medical Physics,2024,41(6):690-701.[doi:DOI:10.3969/j.issn.1005-202X.2024.06.006]
点击复制

基于隐式退化模型的磁共振图像超分辨重建网络()
分享到:

《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]

卷:
41卷
期数:
2024年第6期
页码:
690-701
栏目:
医学影像物理
出版日期:
2024-06-25

文章信息/Info

Title:
Super-resolution reconstruction network based on implicit degradation model for magnetic resonance images
文章编号:
1005-202X(2024)06-0690-12
作者:
刘环宇1郭海鹏1刘晓东2李晗1李君宝1
1.哈尔滨工业大学计算学部信息对抗技术研究所, 黑龙江 哈尔滨 150080; 2.哈尔滨工业大学电子与信息工程学院自动化测试研究所, 黑龙江 哈尔滨 150080
Author(s):
LIU Huanyu1 GUO Haipeng1 LIU Xiaodong2 LI Han1 LI Junbao1
1.Information Countermeasure Technique Institute, Faculty of Computing, Harbin Institute of Technology, Harbin 150080, China 2. Department of Automatic Test and Control, School of Electronics and Information Engineering, Harbin Institute of Technology, Harbin 150080, China
关键词:
脑部磁共振图像超分辨扩散模型
Keywords:
Keywords: brain magnetic resonance imaging super-resolution diffusion model
分类号:
R318;TP183
DOI:
DOI:10.3969/j.issn.1005-202X.2024.06.006
文献标志码:
A
摘要:
对于使用算法提升磁共振(MR)图像分辨率的研究,现有方法多停留在跨尺寸、同尺寸有监督的超分辨算法研究,提出一种基于隐式退化映射模型的MR图像超分辨重建网络SG-Diffusion,通过掩码自编码器隐式建模MR图像的退化过程,减小实验构建数据集与实际场景下MR图像的域差距,并基于隐式退化模型构建样本对,训练得到基于自引导扩散模型的MR图像重建网络,从而实现无监督同尺寸MR图像的空间分辨率提升。在fastMRI数据集的4倍加速采样脑部MR图像超分辨实验结果显示,本文提出的基于隐式退化模型的MR图像超分辨重建网络能够有效提升退化MR图像的空间分辨率,同时与基于显示退化模型的图像退化重建方法相比,本文提出的SG-Diffusion方法具有更好的重建效果。
Abstract:
Abstract: Given that the existing methods of enhancing the resolution of magnetic resonance (MR) images by algorithms mainly focus on cross-size and same-size supervised super-resolution algorithms, a super-resolution reconstruction network (SG-Diffusion) for MR images is proposed based on an implicit degradation mapping model. The degradation process of MR images is implicitly modeled through a masked autoencoder, which reduces the domain gap between the experimental constructed dataset and the actual MR images, and the sample pairs are generated based on implicit degradation model. After training, a MR image reconstruction network based on self-guided diffusion model is obtained to realize the spatial resolution enhancement of unsupervised same-size MR images. The results of super-resolution experiments of 4-fold accelerated sampling brain MR images on fastMRI dataset show that the MR image super-resolution reconstruction network based on implicit degradation model proposed in the study can effectively improve the spatial resolution of degraded MR images, and that compared with the image degradation reconstruction method based on the explicit degradation model, the proposed SG-Diffusion method achieves better reconstruction results.

相似文献/References:

[1]王晓飞,聂生东,王远军,等.改进的K-均值聚类算法及其在脑组织分割中的应用[J].中国医学物理学杂志,2014,31(02):4760.[doi:10.3969/j.issn.1005-202X.2014.02.010]
[2]王晓飞,聂生东,王远军.改进的Brain Extraction Tool算法及其在脑实质分割中的应用[J].中国医学物理学杂志,2016,33(2):113.[doi:10.3969/j.issn.1005-202X.2016.02.002]
 [J].Chinese Journal of Medical Physics,2016,33(6):113.[doi:10.3969/j.issn.1005-202X.2016.02.002]
[3]李均,蒋帆,魏乐,等. 基于磁共振图像构建中国人脑模板[J].中国医学物理学杂志,2017,34(6):614.[doi:DOI:10.3969/j.issn.1005-202X.2017.06.015]
 [J].Chinese Journal of Medical Physics,2017,34(6):614.[doi:DOI:10.3969/j.issn.1005-202X.2017.06.015]

备注/Memo

备注/Memo:
【收稿日期】2024-01-23【基金项目】国家自然科学基金(62271166);哈尔滨工业大学医工理交叉基金(IR2021104)【作者简介】刘环宇,副研究员,硕士生导师,研究方向:机器学习算法、嵌入式智能系统、基于机器学习的网络空间安全,E-mail: liuhuanyu@hit.edu.cn【通信作者】李君宝,博士,博士生导师,研究方向:机器学习算法、人工智能安全、嵌入式智能系统、图像处理,E-mail: lijunbao@hit.edu.cn
更新日期/Last Update: 2024-06-25