Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Aug 5;10(15):e35561.
doi: 10.1016/j.heliyon.2024.e35561. eCollection 2024 Aug 15.

Identifying factors related to mortality of hospitalized COVID-19 patients using machine learning methods

Farzaneh Hamidi  1 , Hadi Hamishehkar  2   3 , Pedram Pirmad Azari Markid  4 , Parvin Sarbakhsh  5
Affiliations

Identifying factors related to mortality of hospitalized COVID-19 patients using machine learning methods

Farzaneh Hamidi et al. Heliyon. .

Abstract

Background: The COVID-19 pandemic has had a profound impact globally, presenting significant social and economic challenges. This study aims to explore the factors affecting mortality among hospitalized COVID-19 patients and construct a machine learning-based model to predict the risk of mortality.

Methods: The study examined COVID-19 patients admitted to Imam Reza Hospital in Tabriz, Iran, between March 2020 and November 2021. The Elastic Net method was employed to identify and rank features associated with mortality risk. Subsequently, an artificial neural network (ANN) model was developed based on these features to predict mortality risk. The performance of the model was evaluated by receiver operating characteristic (ROC) curve analysis.

Results: The study included 706 patients with 96 features, out of them 26 features were identified as crucial predictors of mortality. The ANN model, utilizing 20 of these features, achieved an area under the ROC curve (AUC) of 98.8 %, effectively stratifying patients by mortality risk.

Conclusion: The developed model offers accurate and precipitous mortality risk predictions for COVID-19 patients, enhancing the responsiveness of healthcare systems to high-risk individuals.

Keywords: Artificial neural network; COVID-19; Elastic net; Feature selection; LASSO; Machine learning; Mortality.

PubMed Disclaimer

Conflict of interest statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Figures

Fig. 1
Fig. 1
Tuning parameter (lambda) selection in the Elastic Net model used 10-fold cross-validation based on “lambda.1se” criteria for COVID-19 prognosis.
Fig. 2
Fig. 2
ROC curve for the artificial neural network base on selected features.
Fig. 3
Fig. 3
Box and jitter plots of continues features (A: CRP; B: INR; C: SPO2 first day; D: Urea) according to the mortality outcome.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. WHO Coronavirus Disease (COVID-19) Dashboard DlupC, https://covid19.who.int/. Available from: https://covid19.who.int/.
    1. Shang Y., et al. Scoring systems for predicting mortality for severe patients with COVID-19. EClinicalMedicine. 2020;24 - PMC - PubMed
    1. Amanullah S., Ramesh Shankar R. Healthcare. MDPI; 2020. The impact of COVID-19 on physician burnout globally: a review. - PMC - PubMed
    1. Worby C.J., Chang H.-H. Face mask use in the general population and optimal resource allocation during the COVID-19 pandemic. Nat. Commun. 2020;11(1):4049. - PMC - PubMed
    1. Janke A.T., et al. Analysis of hospital resource availability and COVID‐19 mortality across the United States. J. Hosp. Med. 2021;16(4):211–214. - PMC - PubMed

LinkOut - more resources