MR-Forest: A Deep Decision Framework for False Positive Reduction in Pulmonary Nodule Detection

CORC > 沈阳自动化研究所 > 中国科学院沈阳自动化研究所 > 工业控制网络与系统研究室

	MR-Forest: A Deep Decision Framework for False Positive Reduction in Pulmonary Nodule Detection
	Bi, Yuanguo 6; Bian, Zijian 4; Song CH(宋纯贺)5; Zhao, Hai 6; Zhu, Hongbo 6; Liu, Tong6 ; He, Xuan 3; Yang, Dongxiang 2; Cai, Wei 1
刊名	IEEE Journal of Biomedical and Health Informatics
	2019
页码	1-12
关键词	False positive reduction Spherical surface feature Deep decision ORFs Computer tomography
ISSN号	2168-2194
产权排序	3
英文摘要	With the development of deep learning methods such as convolutional neural network (CNN), the accuracy of automated pulmonary nodule detection has been greatly improved. However, the high computational and storage costs of the large-scale network have been a potential concern for the future widespread clinical application. In this paper, an alternative Multi-ringed (MR)-Forest framework, against the resource-consuming neural networks (NN)-based architectures, has been proposed for false positive reduction in pulmonary nodule detection, which consists of three steps. First, a novel multi-ringed scanning method is used to extract the order ring facets (ORFs) from the surface voxels of the volumetric nodule models; Second, Mesh-LBP and mapping deformation are employed to estimate the texture and shape features. By sliding and resampling the multi-ringed ORFs, feature volumes with different lengths are generated. Finally, the outputs of multilevel are cascaded to predict the candidate class. On 1034 scans merging the dataset from the Affiliated Hospital of Liaoning University of Traditional Chinese Medicine (AH-LUTCM) and the LUNA16 Challenge dataset, our framework performs enough competitiveness than state-of-the-art in false positive reduction task (CPM score of 0.865). Experimental results demonstrate that MR-Forest is a successful substitution to satisfied both resource-consuming and effectiveness for automated pulmonary nodule detection systems. The proposed MR-forest is a general architecture for 3D target detection, it can be easily extended in many other medical imaging analysis tasks, where the growth trend of the targeting object is approximated as a spheroidal expansion.
语种	英语
资助机构	National Key Research and Development Program under Grant 2018YFB1702003 ; National Science Foundation of China under Grant 61806048 ; Open Program of Neusoft Research of Intelligent Healthcare Technology, Co. Ltd. under Grant NRIHTOP1802 ; Fundamental Research Funds for the Central Universities under Grant N180716019
内容类型	期刊论文
源URL	[http://ir.sia.cn/handle/173321/25805]
专题	沈阳自动化研究所_工业控制网络与系统研究室
作者单位	1.Neusoft Institute of Intelligent Healthcare Technology, Co. Ltd., Shenyang, Liaoning China 2.Affiliated Hospital, Liaoning University of Traditional Chinese Medicine, 66473 Shenyang, Liaoning China 3.Sino-Dutch Biomedical and Information Engineering School, Northeastern University, Shenyang, Liaoning China 4.Philips China Investment Co., Ltd., Imaging Clinical Applications Platform, Shanghai China 5.Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, Liaoning China 6.School of Computer Science and Engineering, Northeastern University, Shenyang, Liaoning China
推荐引用方式 GB/T 7714	Bi, Yuanguo,Bian, Zijian,Song CH,et al. MR-Forest: A Deep Decision Framework for False Positive Reduction in Pulmonary Nodule Detection[J]. IEEE Journal of Biomedical and Health Informatics,2019:1-12.
APA	Bi, Yuanguo.,Bian, Zijian.,Song CH.,Zhao, Hai.,Zhu, Hongbo.,...&Cai, Wei.(2019).MR-Forest: A Deep Decision Framework for False Positive Reduction in Pulmonary Nodule Detection.IEEE Journal of Biomedical and Health Informatics,1-12.
MLA	Bi, Yuanguo,et al."MR-Forest: A Deep Decision Framework for False Positive Reduction in Pulmonary Nodule Detection".IEEE Journal of Biomedical and Health Informatics (2019):1-12.