Whole-body tumor segmentation from PET/CT images using a two-stage cascaded neural network with camouflaged object detection mechanisms

doi:10.1002/mp.16438

LZUFE-KMS > 科技处

	Whole-body tumor segmentation from PET/CT images using a two-stage cascaded neural network with camouflaged object detection mechanisms
	He, Jiangping1 ; Zhang, Yangjie 1; Chung, Maggie 2; Wang, Michael 3; Wang, Kun 1; Ma, Yan 1; Ding, Xiaoyang1 ; Li, Qiang 1; Pu, Yonglin 4
	2023
发表期刊	Medical Physics
卷号	50 期号:10 页码:6151-6162
摘要	Background: Whole-body Metabolic Tumor Volume (MTVwb) is an independent prognostic factor for overall survival in lung cancer patients. Automatic segmentation methods have been proposed for MTV calculation. Nevertheless, most of existing methods for patients with lung cancer only segment tumors in the thoracic region. Purpose: In this paper, we present a Two-Stage cascaded neural network integrated with Camouflaged Object Detection mEchanisms (TS-Code-Net) for automatic segmenting tumors from whole-body PET/CT images. Methods: Firstly, tumors are detected from the Maximum Intensity Projection (MIP) images of PET/CT scans, and tumors' approximate localizations along z-axis are identified. Secondly, the segmentations are performed on PET/CT slices that contain tumors identified by the first step. Camouflaged object detection mechanisms are utilized to distinguish the tumors from their surrounding regions that have similar Standard Uptake Values (SUV) and texture appearance. Finally, the TS-Code-Net is trained by minimizing the total loss that incorporates the segmentation accuracy loss and the class imbalance loss. Results: The performance of the TS-Code-Net is tested on a whole-body PET/CT image data-set including 480 Non-Small Cell Lung Cancer (NSCLC) patients with five-fold cross-validation using image segmentation metrics. Our method achieves 0.70, 0.76, and 0.70, for Dice, Sensitivity and Precision, respectively, which demonstrates the superiority of the TS-Code-Net over several existing methods related to metastatic lung cancer segmentation from whole-body PET/CT images. Conclusions: The proposed TS-Code-Net is effective for whole-body tumor segmentation of PET/CT images. Codes for TS-Code-Net are available at: https://github.com/zyj19/TS-Code-Net. © 2023 The Authors. Medical Physics published by Wiley Periodicals LLC on behalf of American Association of Physicists in Medicine.
关键词	Biological organs Diseases Image segmentation Object recognition Textures Tumors Camouflaged object detection CT Image Detection mechanism Lung Cancer Neural-networks Objects detection Tumor segmentation Two-stage cascaded neural network Whole-body Whole-body PET/CT
DOI	10.1002/mp.16438
收录类别	EI ; SCIE
ISSN	0094-2405
语种	英语
WOS研究方向	Radiology, Nuclear Medicine & Medical Imaging
WOS类目	Radiology, Nuclear Medicine & Medical Imaging
WOS记录号	WOS:000979684800001
出版者	John Wiley and Sons Ltd
EI入藏号	20231914066948
EI主题词	Object detection
EI分类号	461.2 Biological Materials and Tissue Engineering ; 723.2 Data Processing and Image Processing
原始文献类型	Article in Press
EISSN	2473-4209
引用统计	被引频次：4[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.lzufe.edu.cn/handle/39EH0E1M/34494
专题	科技处信息工程与人工智能学院党委研究生工作部(学位管理与研究生工作处)
作者单位	1.Department of Electronic Engineering, Lanzhou University of Finance and Economics, Gansu, Lanzhou, China; 2.Department of Radiology, University of California, San Francisco; CA, United States; 3.Department of Pathology, University of California, San Francisco; CA, United States; 4.Department of Radiology, University of Chicago, Chicago; IL, United States
第一作者单位	兰州财经大学
推荐引用方式 GB/T 7714	He, Jiangping,Zhang, Yangjie,Chung, Maggie,et al. Whole-body tumor segmentation from PET/CT images using a two-stage cascaded neural network with camouflaged object detection mechanisms[J]. Medical Physics,2023,50(10):6151-6162.
APA	He, Jiangping.,Zhang, Yangjie.,Chung, Maggie.,Wang, Michael.,Wang, Kun.,...&Pu, Yonglin.(2023).Whole-body tumor segmentation from PET/CT images using a two-stage cascaded neural network with camouflaged object detection mechanisms.Medical Physics,50(10),6151-6162.
MLA	He, Jiangping,et al."Whole-body tumor segmentation from PET/CT images using a two-stage cascaded neural network with camouflaged object detection mechanisms".Medical Physics 50.10(2023):6151-6162.