Which images to label for few-shot medical image analysis?

doi:10.1016/j.media.2024.103200

	Which images to label for few-shot medical image analysis?
	Quan, Quan 1,2; Yao, Qingsong 1,2; Zhu, Heqin 3; Wang, Qiyuan 3; Zhou, S. Kevin 1,3,4,5
	2024-08-01
发表期刊	MEDICAL IMAGE ANALYSIS
ISSN	1361-8415
卷号	96 页码:12
摘要	The success of deep learning methodologies hinges upon the availability of meticulously labeled extensive datasets. However, when dealing with medical images, the annotation process for such abundant training data often necessitates the involvement of experienced radiologists, thereby consuming their limited time resources. In order to alleviate this burden, few-shot learning approaches have been developed, which manage to achieve competitive performance levels with only several labeled images. Nevertheless, a crucial yet previously overlooked problem in few-shot learning is about the selection of template images for annotation before learning, which affects the final performance. In this study, we propose a novel TEmplate Choosing Policy (TECP) that aims to identify and select "the most worthy"images for annotation, particularly within the context of multiple few-shot medical tasks, including landmark detection, anatomy detection, and anatomy segmentation. TECP is composed of four integral components: (1) Self-supervised training, which entails training a pre-existing deep model to extract salient features from radiological images; (2) Alternative proposals for localizing informative regions within the images; and (3) Representative Score Estimation, which involves the evaluation and identification of the most representative samples or templates. (4) Ranking, which rank all candidates and select one with highest representative score. The efficacy of the TECP approach is demonstrated through a series of comprehensive experiments conducted on multiple public datasets. Across all three medical tasks, the utilization of TECP yields noticeable improvements in model performance.
关键词	Few-shot Contrastive learning Augmentation
DOI	10.1016/j.media.2024.103200
收录类别	SCI
语种	英语
资助项目	Natural Science Foundation of China[62271465] ; Suzhou Basic Research Program[SYG202338] ; Open Fund Project of Guangdong Academy of Medical Sciences, China[YKY-KF202206]
WOS研究方向	Computer Science ; Engineering ; Radiology, Nuclear Medicine & Medical Imaging
WOS类目	Computer Science, Artificial Intelligence ; Computer Science, Interdisciplinary Applications ; Engineering, Biomedical ; Radiology, Nuclear Medicine & Medical Imaging
WOS记录号	WOS:001246461100001
出版者	ELSEVIER
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/39925
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Zhou, S. Kevin
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Beijing 100080, Peoples R China 2.Univ Chinese Acad Sci UCAS, Beijing 101408, Peoples R China 3.Univ Sci & Technol China USTC, Sch Biomed Engn, Div Life Sci & Med, Hefei 230026, Peoples R China 4.Suzhou Inst Adv Res, Ctr Med Imaging Robot Analyt Comp Learning MIRACLE, Suzhou 215000, Peoples R China 5.USTC, Key Lab Precis & Intelligent Chem, Hefei 230026, Peoples R China
推荐引用方式 GB/T 7714	Quan, Quan,Yao, Qingsong,Zhu, Heqin,et al. Which images to label for few-shot medical image analysis?[J]. MEDICAL IMAGE ANALYSIS,2024,96:12.
APA	Quan, Quan,Yao, Qingsong,Zhu, Heqin,Wang, Qiyuan,&Zhou, S. Kevin.(2024).Which images to label for few-shot medical image analysis?.MEDICAL IMAGE ANALYSIS,96,12.
MLA	Quan, Quan,et al."Which images to label for few-shot medical image analysis?".MEDICAL IMAGE ANALYSIS 96(2024):12.