A Fast and Refined Cancer Regions Segmentation Framework in Whole-slide Breast Pathological Images

doi:10.1038/s41598-018-37492-9

	A Fast and Refined Cancer Regions Segmentation Framework in Whole-slide Breast Pathological Images
	Guo, Zichao 1; Liu, Hong 1; Ni, Haomiao 1; Wang, Xiangdong 1; Su, Mingming 2,3,4,5; Guo, Wei 6,7; Wang, Kuansong 6,7; Jiang, Taijiao 2,3,4; Qian, Yueliang 1
	2019-01-29
发表期刊	SCIENTIFIC REPORTS
ISSN	2045-2322
卷号	9 页码:10
摘要	Supervised learning methods are commonly applied in medical image analysis. However, the success of these approaches is highly dependent on the availability of large manually detailed annotated dataset. Thus an automatic refined segmentation of whole-slide image (WSI) is significant to alleviate the annotation workload of pathologists. But most of the current ways can only output a rough prediction of lesion areas and consume much time in each slide. In this paper, we propose a fast and refined cancer regions segmentation framework v3_DCNN, which first preselects tumor regions using a classification model Inception-v3 and then employs a semantic segmentation model DCNN for refined segmentation. Our framework can generate a dense likelihood heatmap with the 1/8 side of original WSI in 11.5 minutes on the Camelyon16 dataset, which saves more than one hour for each WSI compared with the initial DCNN model. Experimental results show that our approach achieves a higher FROC score 83.5% with the champion's method of Camelyon16 challenge 80.7%. Based on v3 DCNN model, we further automatically produce heatmap of WSI and extract polygons of lesion regions for doctors, which is very helpful for their pathological diagnosis, detailed annotation and thus contributes to developing a more powerful deep learning model.
DOI	10.1038/s41598-018-37492-9
收录类别	SCI
语种	英语
资助项目	Beijing Natural Science Foundation[4172058] ; Central Public-interest Scientific Institution Basal Research Fund[2016ZX310195] ; Central Public-interest Scientific Institution Basal Research Fund[2017PT31026]
WOS研究方向	Science & Technology - Other Topics
WOS类目	Multidisciplinary Sciences
WOS记录号	WOS:000456955500062
出版者	NATURE PUBLISHING GROUP
引用统计	被引频次：80[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/3443
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Liu, Hong; Wang, Kuansong; Qian, Yueliang
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Beijing Key Lab Mobile Comp & Pervas Device, Beijing 100190, Peoples R China 2.Chinese Acad Med Sci, Res Ctr Big Data Biomed Sci, Inst Basic Med Sci, Beijing 100005, Peoples R China 3.Peking Union Med Coll, Beijing 100005, Peoples R China 4.Suzhou Inst Syst Med, Suzhou 215123, Peoples R China 5.Peking Union Med Coll, Grad Sch, Beijing 100005, Peoples R China 6.Cent S Univ, Xiangya Hosp, Dept Pathol, Changsha 410013, Hunan, Peoples R China 7.Cent S Univ, Sch Basic Med Sci, Dept Pathol, Changsha 410013, Hunan, Peoples R China
推荐引用方式 GB/T 7714	Guo, Zichao,Liu, Hong,Ni, Haomiao,et al. A Fast and Refined Cancer Regions Segmentation Framework in Whole-slide Breast Pathological Images[J]. SCIENTIFIC REPORTS,2019,9:10.
APA	Guo, Zichao.,Liu, Hong.,Ni, Haomiao.,Wang, Xiangdong.,Su, Mingming.,...&Qian, Yueliang.(2019).A Fast and Refined Cancer Regions Segmentation Framework in Whole-slide Breast Pathological Images.SCIENTIFIC REPORTS,9,10.
MLA	Guo, Zichao,et al."A Fast and Refined Cancer Regions Segmentation Framework in Whole-slide Breast Pathological Images".SCIENTIFIC REPORTS 9(2019):10.