TY - GEN
T1 - Chest X-Ray using Brixia and Fractional Inspiratory Oxygen as Severity and Mortality predictor of COVID-19
AU - Rosyid, Alfian Nur
AU - Puspitasari, Arina Dery
AU - Soebakti, Erika
AU - Sensusiati, Anggraini Dwi
AU - Nugraha, Jusak
AU - Amin, Muhammad
N1 - Publisher Copyright:
© 2023 ACM.
PY - 2023/5/12
Y1 - 2023/5/12
N2 - Background: COVID-19 patients often experience pneumonia that worsens conditions and causes death. Lung abnormalities and oxygen needs are usually found in hospitalised COVID-19 patients. Objective: This study aimed to determine the initial Brixia score and FiO2 as predictors of the severity and mortality of COVID-19. Methods: Secondary data were taken from hospitalised COVID-19 patients at the Airlangga University Hospital from March to September 2021. Three reviewers evaluated chest radiographs with a Brixia score, and the inspiratory fraction of oxygen (FiO2) was calculated as 0.21+(4xairflow). Patients with pulmonary TB, asthma, COPD, HIV/AIDS, and pregnant women were excluded from the study. Results: About 72 respondents were 47% male, with an average age of 50.79. The severe group's mean Brixia score and FiO2 were higher than the non-severe group and higher in non-survivor vs survivor groups (p=0.001). Kappa interobserver of Brixia readers was 0.788; p<0.05. The Receiver Operating Characteristics (ROC) obtained the Area Under Curve (AUC) in Brixia for severity =0.874; p=0.000 and mortality =0.759; p=0.000. ROC in FiO2 for severity =0.804; p=0.002 and mortality =0.784; p=0.000. The Brixia score of 10.50 had a sensitivity (Sn)=0.778, specificity (Sp)=0.778 for severity and Sn=0.739, and Sp=0.612 for mortality. The cut-off FiO2 of 47.5% (equivalent to 6 litres per minute of oxygen) had Sn=0.778 and Sp=0.667 for severity and Sn=0.826, and Sp=0.571 for mortality. Conclusion: Initial Chest X-Ray scored by Brixia and the inspired oxygen fraction can be used as initial predictors of the severity and mortality of COVID-19.
AB - Background: COVID-19 patients often experience pneumonia that worsens conditions and causes death. Lung abnormalities and oxygen needs are usually found in hospitalised COVID-19 patients. Objective: This study aimed to determine the initial Brixia score and FiO2 as predictors of the severity and mortality of COVID-19. Methods: Secondary data were taken from hospitalised COVID-19 patients at the Airlangga University Hospital from March to September 2021. Three reviewers evaluated chest radiographs with a Brixia score, and the inspiratory fraction of oxygen (FiO2) was calculated as 0.21+(4xairflow). Patients with pulmonary TB, asthma, COPD, HIV/AIDS, and pregnant women were excluded from the study. Results: About 72 respondents were 47% male, with an average age of 50.79. The severe group's mean Brixia score and FiO2 were higher than the non-severe group and higher in non-survivor vs survivor groups (p=0.001). Kappa interobserver of Brixia readers was 0.788; p<0.05. The Receiver Operating Characteristics (ROC) obtained the Area Under Curve (AUC) in Brixia for severity =0.874; p=0.000 and mortality =0.759; p=0.000. ROC in FiO2 for severity =0.804; p=0.002 and mortality =0.784; p=0.000. The Brixia score of 10.50 had a sensitivity (Sn)=0.778, specificity (Sp)=0.778 for severity and Sn=0.739, and Sp=0.612 for mortality. The cut-off FiO2 of 47.5% (equivalent to 6 litres per minute of oxygen) had Sn=0.778 and Sp=0.667 for severity and Sn=0.826, and Sp=0.571 for mortality. Conclusion: Initial Chest X-Ray scored by Brixia and the inspired oxygen fraction can be used as initial predictors of the severity and mortality of COVID-19.
KW - COVID-19
KW - Imaging
KW - infection disease
KW - oxygen demand
KW - predictor
UR - http://www.scopus.com/inward/record.url?scp=85178072010&partnerID=8YFLogxK
U2 - 10.1145/3608298.3608332
DO - 10.1145/3608298.3608332
M3 - Conference contribution
AN - SCOPUS:85178072010
T3 - ACM International Conference Proceeding Series
SP - 182
EP - 187
BT - ICMHI 2023 - 2023 the 7th International Conference on Medical and Health Informatics
PB - Association for Computing Machinery
T2 - 7th International Conference on Medical and Health Informatics, ICMHI 2023
Y2 - 12 May 2023 through 14 May 2023
ER -