TR-GAN: Multi-Session Future MRI Prediction With Temporal Recurrent Generative Adversarial Network

doi:10.1109/TMI.2022.3151118

CORC > 自动化研究所 > 中国科学院自动化研究所 > 复杂系统管理与控制国家重点实验室 > 先进机器人控制团队

	TR-GAN: Multi-Session Future MRI Prediction With Temporal Recurrent Generative Adversarial Network
	Fan, Chen-Chen 4,5; Peng, Liang 5; Wang, Tian 3; Yang, Hongjun 5; Zhou, Xiao-Hu 5; Ni, Zhen-Liang 4,5; Wang, Guan'an 4,5; Chen, Sheng 4,5; Zhou, Yan-Jie 4,5; Hou, Zeng-Guang 1,2,4,5
刊名	IEEE TRANSACTIONS ON MEDICAL IMAGING
	2022-08-01
卷号	41 期号:8 页码:1925-1937
关键词	Magnetic resonance imaging Generative adversarial networks Task analysis Three-dimensional displays Training Generators Data models Alzheimer's disease magnetic resonance imaging generative adversarial network
ISSN号	0278-0062
DOI	10.1109/TMI.2022.3151118
通讯作者	Hou, Zeng-Guang(zengguang.hou@ia.ac.cn)
英文摘要	Magnetic Resonance Imaging (MRI) has been proven to be an efficient way to diagnose Alzheimer's disease (AD). Recent dramatic progress on deep learning greatly promotes the MRI analysis based on data-driven CNN methods using a large-scale longitudinal MRI dataset. However, most of the existing MRI datasets are fragmented due to unexpected quits of volunteers. To tackle this problem, we propose a novel Temporal Recurrent Generative Adversarial Network (TR-GAN) to complete missing sessions of MRI datasets. Unlike existing GAN-based methods, which either fail to generate future sessions or only generate fixed-length sessions, TR-GAN takes all past sessions to recurrently and smoothly generate future ones with variant length. Specifically, TR-GAN adopts recurrent connection to deal with variant input sequence length and flexibly generate future variant sessions. Besides, we also design a multiple scale & location (MSL) module and a SWAP module to encourage the model to better focus on detailed information, which helps to generate high-quality MRI data. Compared with other popular GAN architectures, TR-GAN achieved the best performance in all evaluation metrics of two datasets. After expanding the Whole MRI dataset, the balanced accuracy of AD vs. cognitively normal (CN) vs. mild cognitive impairment (MCI) and stable MCI vs. progressive MCI classification can be increased by 3.61% and 4.00%, respectively.
资助项目	National Key Research and Development Program of China[2018YFC2001700] ; National Natural Science Foundation of China[61720106012] ; National Natural Science Foundation of China[U1913601] ; National Natural Science Foundation of China[62073319] ; National Natural Science Foundation of China[62003343] ; National Natural Science Foundation of China[U20A20224] ; Beijing Natural Science Foundation[L172050] ; Beijing SciTech Program[Z211100007921021] ; Youth Innovation Promotion Association of Chinese Academy of Sciences[2020140] ; Strategic Priority Research Program of Chinese Academy of Science[XDB32040000] ; Alzheimer's Disease Neuroimaging Initiative (ADNI) (National Institutes of Health)[U01 AG024904] ; Department of Defense (DOD) ADNI[W81XWH-12-2-0012]
WOS关键词	ALZHEIMERS-DISEASE ; IMAGE
WOS研究方向	Computer Science ; Engineering ; Imaging Science & Photographic Technology ; Radiology, Nuclear Medicine & Medical Imaging
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000837269000003
资助机构	National Key Research and Development Program of China ; National Natural Science Foundation of China ; Beijing Natural Science Foundation ; Beijing SciTech Program ; Youth Innovation Promotion Association of Chinese Academy of Sciences ; Strategic Priority Research Program of Chinese Academy of Science ; Alzheimer's Disease Neuroimaging Initiative (ADNI) (National Institutes of Health) ; Department of Defense (DOD) ADNI
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/49894]
专题	自动化研究所_复杂系统管理与控制国家重点实验室_先进机器人控制团队
通讯作者	Hou, Zeng-Guang
作者单位	1.Macau Univ Sci & Technol, CASIA MUST Joint Lab Intelligence Sci & Technol, Inst Syst Engn, Macau 999078, Peoples R China 2.CAS Ctr Excellence Brain Sci & Intelligence Techn, Beijing 100190, Peoples R China 3.Commun Univ China, Neurosci & Intelligent Media Inst, Beijing 100024, Peoples R China 4.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China 5.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Fan, Chen-Chen,Peng, Liang,Wang, Tian,et al. TR-GAN: Multi-Session Future MRI Prediction With Temporal Recurrent Generative Adversarial Network[J]. IEEE TRANSACTIONS ON MEDICAL IMAGING,2022,41(8):1925-1937.
APA	Fan, Chen-Chen.,Peng, Liang.,Wang, Tian.,Yang, Hongjun.,Zhou, Xiao-Hu.,...&Hou, Zeng-Guang.(2022).TR-GAN: Multi-Session Future MRI Prediction With Temporal Recurrent Generative Adversarial Network.IEEE TRANSACTIONS ON MEDICAL IMAGING,41(8),1925-1937.
MLA	Fan, Chen-Chen,et al."TR-GAN: Multi-Session Future MRI Prediction With Temporal Recurrent Generative Adversarial Network".IEEE TRANSACTIONS ON MEDICAL IMAGING 41.8(2022):1925-1937.