A Multi-Center Study of COVID-19 Patient Prognosis Using Deep Learning-based CT Image Analysis and Electronic Health Records

Autor: Ittai Dayan, Kuang Gong, Rosa Babaei, Fatemeh Homayounieh, Alessandro Carriero, Won Young Tak, Subba R. Digumarthy, Varun Buch, Mahsa Masjedi, Soo-Young Park, Luca Saba, Min Kyu Kang, Quanzheng Li, Nir Neumark, Hamid Reza Talari, Hui Ren, Ning Guo, Hadi Karimi Mobin, Dufan Wu, Yu Rim Lee, Mannudeep K. Kalra, Bernardo Bizzo, Jiahui Guan, Shadi Ebrahimian, Chiara Arru, Jung Gil Park, Kyungsang Kim
Jazyk: angličtina
Rok vydání: 2021
Předmět:
Disease
Health records
030218 nuclear medicine & medical imaging
0302 clinical medicine
COVID-19
Coronavirus disease of 2019
Medicine
Electronic health records
Lung
GLM
Generalized linear model
SOFA
Sequential Organ Failure Assessment
WBC
White blood cell
Hgb
Hemoglobin
General Medicine
Prognosis
RT-PCR
Reverse-transcription polymerase chain reaction
ICU
Intensive care unit
GPU
Graphics processing unit
Radiology Nuclear Medicine and imaging
PLT
Platelet
HU
Hounsfield unit
030220 oncology & carcinogenesis
Cohort
MV
Mechanical ventilation
SpO2
Oxygen saturation
CT
Computed tomography
hs-CRP
High-sensitivity C-reactive protein
Radiology
Research Article
Prognosis prediction
medicine.medical_specialty
IRB
Institutional Review Board
Coronavirus disease 2019 (COVID-19)
CR
Consolidation ratio
Vital signs
TOR
Total opacity ratio
03 medical and health sciences
MODS
Multiple Organ Dysfunction Score
Computed Tomography
Deep Learning
GGO
Ground-glass opacity
Humans
Radiology
Nuclear Medicine and imaging
EHR
Electronic health records
LDH
Lactate dehydrogenase
business.industry
SARS-CoV-2
CI
Confidence interval
Deep learning
COVID-19
AUC
Area under the curve
Multi center study
Artificial intelligence
business
Tomography
X-Ray Computed
ESR
Erythrocyte sedimentation rate
Zdroj: European Journal of Radiology
ISSN: 1872-7727
0720-048X
Popis: Purpose As of August 30th, there were in total 25.1 million confirmed cases and 845 thousand deaths caused by coronavirus disease of 2019 (COVID-19) worldwide. With overwhelming demands on medical resources, patient stratification based on their risks is essential. In this multi-center study, we built prognosis models to predict severity outcomes, combining patients’ electronic health records (EHR), which included vital signs and laboratory data, with deep learning- and CT-based severity prediction. Method We first developed a CT segmentation network using datasets from multiple institutions worldwide. Two biomarkers were extracted from the CT images: total opacity ratio (TOR) and consolidation ratio (CR). After obtaining TOR and CR, further prognosis analysis was conducted on datasets from INSTITUTE-1, INSTITUTE-2 and INSTITUTE-3. For each data cohort, generalized linear model (GLM) was applied for prognosis prediction. Results For the deep learning model, the correlation coefficient of the network prediction and manual segmentation was 0.755, 0.919, and 0.824 for the three cohorts, respectively. The AUC (95 % CI) of the final prognosis models was 0.85(0.77,0.92), 0.93(0.87,0.98), and 0.86(0.75,0.94) for INSTITUTE-1, INSTITUTE-2 and INSTITUTE-3 cohorts, respectively. Either TOR or CR exist in all three final prognosis models. Age, white blood cell (WBC), and platelet (PLT) were chosen predictors in two cohorts. Oxygen saturation (SpO2) was a chosen predictor in one cohort. Conclusion The developed deep learning method can segment lung infection regions. Prognosis results indicated that age, SpO2, CT biomarkers, PLT, and WBC were the most important prognostic predictors of COVID-19 in our prognosis model.
Databáze: OpenAIRE
Externí odkaz: