基于深度学习的MR图像重建算法进展  

Advances in deep learning-based magnetic resonance image reconstruction algorithms

作　　者：魏海宁 李睿[1] WEI Haining;LI Rui(Center for Biomedical Imaging Research,Tsinghua University,Beijing 100084,China)

机构地区：[1]清华大学生物医学影像研究中心,北京100084

出　　处：《中国研究型医院》2022年第3期13-19,共7页Chinese Research Hospitals

基　　金：国家自然科学基金(81971604);北京自然科学基金(L192013);清华大学精准医学研究计划(10001020104)。

摘　　要：MR是一种临床常用的、无创成像技术。由于其采集、成像时间长使其推广应用受到限制,故为了减少扫描时间,基于信号处理的不完整K空间重建技术成为新的发展趋势。随着深度学习在医学成像应用上的成功,基于神经网络的降采样MR图像重建算法受到了广泛的关注,其重建结果及效率优于传统的压缩感知(CS)等重建方法。在此就重建算法的发展历程,对基于深度学习的MR重建算法从模型、数据流、训练方面进行系统阐述。Magnetic resonance(MR)is a commonly used,non-invasive imaging technique in clinical practice.However,the long acquisition time limits its widespread application.Therefore,to reduce the scanning time,incomplete K-space reconstruction technology based on signal processing has become a new development trend.With the success of deep learning in medical imaging applications,neural network-based downsampling MR image reconstruction algorithms have received extensive attention,and their reconstruction results and efficiency are better than traditional reconstruction methods such as compressed sensing(CS).In this survey,the development history of reconstruction algorithms is described systematically from the model,data flow,and training side for deep learning-based MR reconstruction algorithms.

关 键 词：磁共振成像 深度学习 降采样图像重建 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP391.41[自动化与计算机技术—控制科学与工程] R445.2[医药卫生—影像医学与核医学]