Research on An Early Warning System of Coronavirus Disease 2019 Based on
Artificial Intelligence in Radiology Department
ZHANG Kai1
, LIU Xiumin1
, CHEN Yuhuan2
, TAN Jia1
, SONG Tingni1
, LI Zhenlin1
1. Department of Radiology, West China Hospital of Sichuan University, Chengdu Sichuan 610041, China;
2. Beijing Infervision Technology Co. LTD, Beijing 100025, China
Abstract：Objective To build an early warning system for screening and diagnosing coronavirus disease 2019 (COVID-19) based
on artificial intelligence in radiology department and explore its application effect. Methods The image-assisted diagnostic model
for COVID-19 based on deep learning technology was applied to the imaging technician inspection and physician diagnosis
workflow. Meanwhile, the emergency mechanism of the department was started to make further clinical decision and treatment
for suspected patients. A total of 210 chest CT data of our hospital (including 60 COVID-19 diagnostic data) were retrospectively
included as model test data, and a total of 1200 chest CT data were prospectively included to evaluate the practical application
effect of the warning system. Results The mean detection time of the model was (105.80±48.50) s and the overall treatment
time for suspected COVID-19 patients from early warning to docking with relevant departments in hospital was (5±3) min.
The sensitivity and specificity of the early warning system for screening and diagnosing COVID-19 were 100% and 82.67%,
respectively. Conclusion The artificial intelligence-based COVID-19 early warning system in radiology department can quickly and
accurately screen COVID-19 patients and provide timely intelligent early warning tips, which can help reduce the waiting time of
COVID-19 patients in radiology department and optimize the treatment process.
张凯1
，刘秀民1
，陈玉环2
，谭佳1
，宋婷妮1
，李真林1. 基于人工智能的新型冠状病毒肺炎放射科预警系统研究[J]. 中国医疗设备, 2020, 35(6): 63-66.
ZHANG Kai1
, LIU Xiumin1
, CHEN Yuhuan2
, TAN Jia1
, SONG Tingni1
, LI Zhenlin1. Research on An Early Warning System of Coronavirus Disease 2019 Based on
Artificial Intelligence in Radiology Department. China Medical Devices, 2020, 35(6): 63-66.
