Chest Radiographs Using a Context-Fusion Convolution Neural Network (CNN): Can It Distinguish the Etiology of Community-Acquired Pneumonia (CAP) in Children?

doi:10.1007/s10278-021-00543-1

	Chest Radiographs Using a Context-Fusion Convolution Neural Network (CNN): Can It Distinguish the Etiology of Community-Acquired Pneumonia (CAP) in Children?
	Hu, Shasha 1; Zhu, Yongbei 2; Dong, Di2,3 ; Wang, Bei 1; Zhou, Zuofu 5; Wang, Chi 1; Tian, Jie 2,4; Peng, Yun 1
发表期刊	JOURNAL OF DIGITAL IMAGING
ISSN	0897-1889
	2022-05-18
页码	12
通讯作者	Tian, Jie(tian@ieee.org) ; Peng, Yun(ppengyun@yahoo.com)
摘要	Clinical symptoms and inflammatory markers cannot reliably distinguish the etiology of CAP, and chest radiographs have abundant information related with CAP. Hence, we developed a context-fusion convolution neural network (CNN) to explore the application of chest radiographs to distinguish the etiology of CAP in children. This retrospective study included 1769 cases of pediatric pneumonia (viral pneumonia, n = 487; bacterial pneumonia, n = 496; and mycoplasma pneumonia, n = 786). The chest radiographs of the first examination, C-reactive protein (CRP), and white blood cell (WBC) were collected for analysis. All patients were stochastically divided into training, validation, and test cohorts in a 7:1:2 ratio. Automatic lung segmentation and hand-crafted pneumonia lesion segmentation were performed, from which three image-based models including a full-lung model, a local-lesion model, and a context-fusion model were built; two clinical characteristics were used to build a clinical model, while a logistic regression model combined the best CNN model and two clinical characteristics. Our experiments showed that the context-fusion model which integrated the features of the full-lung and local-lesion had better performance than the full-lung model and local-lesion model. The context-fusion model had area under curves of 0.86, 0.88, and 0.93 in identifying viral, bacterial, and mycoplasma pneumonia on the test cohort respectively. The addition of clinical characteristics to the context-fusion model obtained slight improvement. Mycoplasma pneumonia was more easily identified compared with the other two types. Using chest radiographs, we developed a context-fusion CNN model with good performance for noninvasively diagnosing the etiology of community-acquired pneumonia in children, which would help improve early diagnosis and treatment.
关键词	Chest radiographs Convolution neural network Community-acquired pneumonia Pediatric Etiology
DOI	10.1007/s10278-021-00543-1
关键词[WOS]	INFLAMMATORY MARKERS ; CLINICAL-FEATURES ; DIAGNOSIS ; UTILITY
收录类别	SCI
语种	英语
资助项目	Special Fund of the Pediatric Medical Coordinated Development Center of Beijing Hospitals Authority[XTCX201814] ; National Key R&D Program of China[2017YFA0205200] ; National Natural Science Foundation of China[82022036] ; National Natural Science Foundation of China[91959130] ; National Natural Science Foundation of China[81971776] ; National Natural Science Foundation of China[81771924] ; National Natural Science Foundation of China[81930053] ; Beijing Natural Science Foundation[L182061] ; Youth Innovation Promotion Association CAS[2017175]
项目资助者	Special Fund of the Pediatric Medical Coordinated Development Center of Beijing Hospitals Authority ; National Key R&D Program of China ; National Natural Science Foundation of China ; Beijing Natural Science Foundation ; Youth Innovation Promotion Association CAS
WOS研究方向	Radiology, Nuclear Medicine & Medical Imaging
WOS类目	Radiology, Nuclear Medicine & Medical Imaging
WOS记录号	WOS:000797262700001
出版者	SPRINGER
引用统计	被引频次：2[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/49462
专题	中国科学院分子影像重点实验室
通讯作者	Tian, Jie; Peng, Yun
作者单位	1.Capital Med Univ, Beijing Childrens Hosp, Natl Ctr Children Hlth, Dept Radiol, Beijing 100045, Peoples R China 2.Chinese Acad Sci, Beijing Key Lab Mol Imaging Inst Automat, State Key Lab Management & Control Complex Syst, CAS Key Lab Mol Imaging, Beijing 100190, Peoples R China 3.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China 4.Beihang Univ, Beijing Adv Innovat Ctr Big Data Based Precis Med, Beijing 100191, Peoples R China 5.Fujian Med Univ, Fujian Prov Matern & Childrens Hosp, Dept Radiol, Fuzhou 350000, Peoples R China
推荐引用方式 GB/T 7714	Hu, Shasha,Zhu, Yongbei,Dong, Di,et al. Chest Radiographs Using a Context-Fusion Convolution Neural Network (CNN): Can It Distinguish the Etiology of Community-Acquired Pneumonia (CAP) in Children?[J]. JOURNAL OF DIGITAL IMAGING,2022:12.
APA	Hu, Shasha.,Zhu, Yongbei.,Dong, Di.,Wang, Bei.,Zhou, Zuofu.,...&Peng, Yun.(2022).Chest Radiographs Using a Context-Fusion Convolution Neural Network (CNN): Can It Distinguish the Etiology of Community-Acquired Pneumonia (CAP) in Children?.JOURNAL OF DIGITAL IMAGING,12.
MLA	Hu, Shasha,et al."Chest Radiographs Using a Context-Fusion Convolution Neural Network (CNN): Can It Distinguish the Etiology of Community-Acquired Pneumonia (CAP) in Children?".JOURNAL OF DIGITAL IMAGING (2022):12.