TY  - JOUR
AU  - Lee, Hyun Woo
AU  - Yang, Hyun Jun
AU  - Kim, Hyungjin
AU  - Kim, Ue-Hwan
AU  - Kim, Dong Hyun
AU  - Yoon, Soon Ho
AU  - Ham, Soo-Youn
AU  - Nam, Bo Da
AU  - Chae, Kum Ju
AU  - Lee, Dabee
AU  - Yoo, Jin Young
AU  - Bak, So Hyeon
AU  - Kim, Jin Young
AU  - Kim, Jin Hwan
AU  - Kim, Ki Beom
AU  - Jung, Jung Im
AU  - Lim, Jae-Kwang
AU  - Lee, Jong Eun
AU  - Chung, Myung Jin
AU  - Lee, Young Kyung
AU  - Kim, Young Seon
AU  - Lee, Sang Min
AU  - Kwon, Woocheol
AU  - Park, Chang Min
AU  - Kim, Yun-Hyeon
AU  - Jeong, Yeon Joo
AU  - Jin, Kwang Nam
AU  - Goo, Jin Mo
PY  - 2023
DA  - 2023/2/16
TI  - Deep Learning With Chest Radiographs for Making Prognoses in Patients With COVID-19: Retrospective Cohort Study
JO  - J Med Internet Res
SP  - e42717
VL  - 25
KW  - COVID-19
KW  - deep learning
KW  - artificial intelligence
KW  - radiography, thoracic
KW  - prognosis
KW  - AI model
KW  - prediction model
KW  - clinical outcome
KW  - medical imaging
KW  - machine learning
AB  - Background: An artificial intelligence (AI) model using chest radiography (CXR) may provide good performance in making prognoses for COVID-19. Objective: We aimed to develop and validate a prediction model using CXR based on an AI model and clinical variables to predict clinical outcomes in patients with COVID-19. Methods: This retrospective longitudinal study included patients hospitalized for COVID-19 at multiple COVID-19 medical centers between February 2020 and October 2020. Patients at Boramae Medical Center were randomly classified into training, validation, and internal testing sets (at a ratio of 8:1:1, respectively). An AI model using initial CXR images as input, a logistic regression model using clinical information, and a combined model using the output of the AI model (as CXR score) and clinical information were developed and trained to predict hospital length of stay (LOS) ≤2 weeks, need for oxygen supplementation, and acute respiratory distress syndrome (ARDS). The models were externally validated in the Korean Imaging Cohort of COVID-19 data set for discrimination and calibration. Results: The AI model using CXR and the logistic regression model using clinical variables were suboptimal to predict hospital LOS ≤2 weeks or the need for oxygen supplementation but performed acceptably in the prediction of ARDS (AI model area under the curve [AUC] 0.782, 95% CI 0.720-0.845; logistic regression model AUC 0.878, 95% CI 0.838-0.919). The combined model performed better in predicting the need for oxygen supplementation (AUC 0.704, 95% CI 0.646-0.762) and ARDS (AUC 0.890, 95% CI 0.853-0.928) compared to the CXR score alone. Both the AI and combined models showed good calibration for predicting ARDS (P=.079 and P=.859). Conclusions: The combined prediction model, comprising the CXR score and clinical information, was externally validated as having acceptable performance in predicting severe illness and excellent performance in predicting ARDS in patients with COVID-19. 
SN  - 1438-8871
UR  - https://www.jmir.org/2023/1/e42717
UR  - https://doi.org/10.2196/42717
UR  - http://www.ncbi.nlm.nih.gov/pubmed/36795468
DO  - 10.2196/42717
ID  - info:doi/10.2196/42717
ER  -