Self-Supervised Arbitrary Scale Super-Resolution Framework for Anisotropic MRI

doi:10.1109/ISBI53787.2023.10230678

	Self-Supervised Arbitrary Scale Super-Resolution Framework for Anisotropic MRI
	Haonan Zhang1 ; Yuhan Zhang1 ; Qing Wu1 ; Jiangjie Wu1 ; Zhiming Zhen 2; Feng Shi 3; Jianmin Yuan 4; Hongjiang Wei 5; Chen Liu 2; Yuyao Zhang1
	2023-04-18
会议录名称	2023 IEEE 20TH INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING (ISBI)
ISSN	1945-7928
卷号	2023-April
发表状态	已发表
DOI	10.1109/ISBI53787.2023.10230678
摘要	In this paper, we propose an efficient self-supervised arbitrary-scale super-resolution (SR) framework to reconstruct isotropic magnetic resonance (MR) images from anisotropic MRI inputs without involving external training data. The proposed framework builds a training dataset using "in-the-wild"anisotropic MR volumes with arbitrary image resolution. We then formulate the 3D volume SR task as a SR problem for 2D image slices. The anisotropic volume's high-resolution (HR) plane is used to build the HR-LR image pairs for model training. We further adapt the implicit neural representation (INR) network to implement the 2D arbitrary-scale image SR model. Finally, we leverage the well-trained proposed model to up-sample the 2D LR plane extracted from the anisotropic MR volumes to their HR views. The isotropic MR volumes thus can be reconstructed by stacking and averaging the generated HR slices. Our proposed framework has two major advantages: (1) It only involves the arbitrary-resolution anisotropic MR volumes, which greatly improves the model practicality in real MR imaging scenarios (e.g., clinical brain image acquisition); (2) The INR-based SR model enables arbitrary-scale image SR from the arbitrary-resolution input image, which significantly improves model training efficiency. We perform experiments on a simulated public adult brain dataset and a real collected 7T brain dataset. The results indicate that our current framework greatly outperforms two well-known self-supervised models for anisotropic MR image SR tasks. © 2023 IEEE.
会议录编者/会议主办者	Flywheel ; Kitware ; Siemens Healthineers ; UCLouvain
关键词	MRI super-resolution self-supervised learning arbitrary scales implicit neural representation
会议名称	20th IEEE International Symposium on Biomedical Imaging, ISBI 2023
会议地点	Cartagena, Colombia
会议日期	18-21 April 2023
URL	查看原文
收录类别	EI
语种	英语
出版者	IEEE Computer Society
EI入藏号	20233914806126
EI主题词	Magnetic resonance
EISSN	1945-8452
EI分类号	461.1 Biomedical Engineering ; 701.2 Magnetism: Basic Concepts and Phenomena ; 746 Imaging Techniques ; 931.2 Physical Properties of Gases, Liquids and Solids
原始文献类型	Conference article (CA)
来源库	IEEE
引用统计	正在获取...
文献类型	会议论文
条目标识符	https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/333438
专题	信息科学与技术学院信息科学与技术学院_硕士生信息科学与技术学院_博士生信息科学与技术学院_PI研究组_张玉瑶组
作者单位	1.School of Information Science and Technology, ShanghaiTech University, Shanghai, China 2.Department of Radiology, Southwest Hospital, Army Medical University (Third Military Medical University), Chongqing, China 3.Shanghai United Imaging Intelligence Co., Ltd., Shanghai, China 4.Central Research Institute, UIH Group, Shanghai, China 5.School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China
第一作者单位	信息科学与技术学院
第一作者的第一单位	信息科学与技术学院
推荐引用方式 GB/T 7714	Haonan Zhang,Yuhan Zhang,Qing Wu,et al. Self-Supervised Arbitrary Scale Super-Resolution Framework for Anisotropic MRI[C]//Flywheel, Kitware, Siemens Healthineers, UCLouvain:IEEE Computer Society,2023.