Multi-scale deformable transformer for multi-contrast knee MRI super-resolution

doi:10.1016/j.bspc.2022.104154

CORC > 自动化研究所 > 中国科学院自动化研究所 > 精密感知与控制研究中心 > 人工智能与机器学习

	Multi-scale deformable transformer for multi-contrast knee MRI super-resolution
	Zou, Beiji 1,3; Ji, Zexin 1,3; Zhu, Chengzhang 1,2,3; Dai, Yulan 1,3; Zhang, Wensheng 4; Kui, Xiaoyan 1,3
刊名	BIOMEDICAL SIGNAL PROCESSING AND CONTROL
	2023
卷号	79 页码:11
关键词	Magnetic resonance imaging Multi-contrast reconstruction Super-resolution Transformer Convolutional neural network
ISSN号	1746-8094
DOI	10.1016/j.bspc.2022.104154
通讯作者	Kui, Xiaoyan(xykui@csu.edu.cn)
英文摘要	Background and objective: Magnetic resonance imaging can present the precise anatomic structure in clinical applications. Nevertheless, due to the limited scanning equipment cost, scanning time and so on, high -resolution knee MR images are difficult to obtain. So the super-resolution technique is developed to improve the image quality. Unfortunately, conventional CNN-based methods cannot explicitly learn the long-range dependencies in images and simply integrate the auxiliary contrast without considering the characteristics of medical images. To tackle this issue, our approach aims to adaptively capture and fuse the significant auxiliary information of the multi-contrast images to improve the knee magnetic resonance image quality.Methods: We propose a multi-scale deformable transformer network (MSDT) for multi-contrast knee magnetic resonance imaging super-resolution. First, we aggregate multi-scale patch embedding from the multi-contrast knee MR images to effectively preserve the local contextual details and global structure information. Then, the deformable transformer architecture is designed to learn the data-dependent sparse attention of the knee MR image, which can adaptively obtain the high-frequency foreground details according to the image content.Results: The proposed method is evaluated on the fastMRI dataset under 2x and 4x enlargements. Our MSDT achieves higher PSNR of 31.98 and SSIM of 0.713 at 2x upsampling factor and PSNR of 30.38 and SSIM of 0.615 at 4x upsampling factor. Moreover, our method can generate clear tissue structures and fine details.Conclusions: The experimental results show superior performance in comparison to the state-of-the-art super -resolution methods. This indicates that the MSDT can effectively reconstruct the high-quality knee MR images.
资助项目	National Key R&D Program of China[2018AAA0102100] ; National Natural Science Foundation of China[62177047] ; Key Research and Development Program of Hunan Province[2022SK2054] ; Scientific and Technological Innovation Leading Plan of High-tech Industry of Hunan Province[2020GK2021] ; Interna-tional Science and Technology Innovation Joint Base of Machine Vision and Medical Image Processing in Hunan Province[2021CB1013] ; Natural Science Foundation of Hunan Province[2022JJ30762] ; 111 project[B18059]
WOS关键词	IMAGE SUPERRESOLUTION ; ALGORITHM ; NETWORK
WOS研究方向	Engineering
语种	英语
出版者	ELSEVIER SCI LTD
WOS记录号	WOS:000862722500008
资助机构	National Key R&D Program of China ; National Natural Science Foundation of China ; Key Research and Development Program of Hunan Province ; Scientific and Technological Innovation Leading Plan of High-tech Industry of Hunan Province ; Interna-tional Science and Technology Innovation Joint Base of Machine Vision and Medical Image Processing in Hunan Province ; Natural Science Foundation of Hunan Province ; 111 project
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/50413]
专题	精密感知与控制研究中心_人工智能与机器学习
通讯作者	Kui, Xiaoyan
作者单位	1.Cent South Univ, Sch Comp Sci & Engn, Changsha 410083, Peoples R China 2.Cent South Univ, Coll Literature & Journalism, Changsha 410083, Peoples R China 3.Cent South Univ, Hunan Engn Res Ctr Machine Vis & Intelligent Med, Changsha 410083, Peoples R China 4.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Zou, Beiji,Ji, Zexin,Zhu, Chengzhang,et al. Multi-scale deformable transformer for multi-contrast knee MRI super-resolution[J]. BIOMEDICAL SIGNAL PROCESSING AND CONTROL,2023,79:11.
APA	Zou, Beiji,Ji, Zexin,Zhu, Chengzhang,Dai, Yulan,Zhang, Wensheng,&Kui, Xiaoyan.(2023).Multi-scale deformable transformer for multi-contrast knee MRI super-resolution.BIOMEDICAL SIGNAL PROCESSING AND CONTROL,79,11.
MLA	Zou, Beiji,et al."Multi-scale deformable transformer for multi-contrast knee MRI super-resolution".BIOMEDICAL SIGNAL PROCESSING AND CONTROL 79(2023):11.