TY - JOUR
T1 - Finding of the factors affecting the severity of COVID-19 based on mathematical models
AU - Qu, Jiahao
AU - Sumali, Brian
AU - Lee, Ho
AU - Terai, Hideki
AU - Ishii, Makoto
AU - Fukunaga, Koichi
AU - Mitsukura, Yasue
AU - Nishimura, Toshihiko
N1 - Funding Information:
Ethics and funding. This study was approved by the Ethics Committee of Keio University and performed in accordance with the Declaration of Helsinki. The study protocol was approved by the Ethics Review Board of Keio University (approval ID 20,200,055). The study was funded by Keio Research Global Institute (grant number: UMIN000041186).
Publisher Copyright:
© 2021, The Author(s).
PY - 2021/12
Y1 - 2021/12
N2 - Since 2019, a large number of people worldwide have been infected with severe acute respiratory syndrome coronavirus 2. Among those infected, a limited number develop severe coronavirus disease 2019 (COVID-19), which generally has an acute onset. The treatment of patients with severe COVID-19 is challenging. To optimize disease prognosis and effectively utilize medical resources, proactive measures must be adopted for patients at risk of developing severe COVID-19. We analyzed the data of COVID-19 patients from seven medical institutions in Tokyo and used mathematical modeling of patient blood test results to quantify and compare the predictive ability of multiple prognostic indicators for the development of severe COVID-19. A machine learning logistic regression model was used to analyze the blood test results of 300 patients. Due to the limited data set, the size of the training group was constantly adjusted to ensure that the results of machine learning were effective (e.g., recognition rate of disease severity > 80%). Lymphocyte count, hemoglobin, and ferritin levels were the best prognostic indicators of severe COVID-19. The mathematical model developed in this study enables prediction and classification of COVID-19 severity.
AB - Since 2019, a large number of people worldwide have been infected with severe acute respiratory syndrome coronavirus 2. Among those infected, a limited number develop severe coronavirus disease 2019 (COVID-19), which generally has an acute onset. The treatment of patients with severe COVID-19 is challenging. To optimize disease prognosis and effectively utilize medical resources, proactive measures must be adopted for patients at risk of developing severe COVID-19. We analyzed the data of COVID-19 patients from seven medical institutions in Tokyo and used mathematical modeling of patient blood test results to quantify and compare the predictive ability of multiple prognostic indicators for the development of severe COVID-19. A machine learning logistic regression model was used to analyze the blood test results of 300 patients. Due to the limited data set, the size of the training group was constantly adjusted to ensure that the results of machine learning were effective (e.g., recognition rate of disease severity > 80%). Lymphocyte count, hemoglobin, and ferritin levels were the best prognostic indicators of severe COVID-19. The mathematical model developed in this study enables prediction and classification of COVID-19 severity.
UR - http://www.scopus.com/inward/record.url?scp=85121528181&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85121528181&partnerID=8YFLogxK
U2 - 10.1038/s41598-021-03632-x
DO - 10.1038/s41598-021-03632-x
M3 - Article
C2 - 34930966
AN - SCOPUS:85121528181
VL - 11
JO - Scientific Reports
JF - Scientific Reports
SN - 2045-2322
IS - 1
M1 - 24224
ER -