Artificial Intelligence for Predicting Mortality in Hospitalized COVID-19 Patients

Overview

Journal Digit Health

Specialty Medical Informatics

Date 2024 Oct 7

PMID 39372811

Authors

Igor N Korsakov

Tatiana L Karonova

Arina A Mikhaylova

Alexander A Loboda

Alyona T Chernikova

Anna G Mikheeva

Marina V Sharypova

Alexandra O Konradi

Evgeny V Shlyakhto

Affiliations

Soon will be listed here.

Abstract

Background: The global demographic situation has been significantly impacted by the COVID-19 pandemic. The objective of this study was to develop a model that predicts the risk of COVID-associated mortality using clinical and laboratory data collected within 72 h of hospital admission.

Materials And Methods: A total of 3024 subjects with PCR-confirmed COVID-19 were admitted to Almazov National Research Medical Center between May 2020 and August 2021. Among them, 6.25% ( = 189) of patients had a fatal outcome. Five machine learning models and the Boruta-SHAP feature selection method were utilized to assess the risk of mortality during COVID-19 hospitalization.

Results: All methods demonstrated high efficacy, with ROC AUC (Receiver Operating Characteristic Area Under the Curve) values exceeding 80%. The selected Boruta-SHAP features, when incorporated into the random forest model, achieved an ROC AUC of 93.1% in the validation.

Conclusion: Throughout the study, close collaboration with healthcare professionals ensured that the developed tool met their practical needs. The success of our model validates the potential of machine learning techniques as decision support systems in clinical practice.

References

Chen R, Chen J, Yang S, Luo S, Xiao Z, Lu L . Prediction of prognosis in COVID-19 patients using machine learning: A systematic review and meta-analysis. Int J Med Inform. 2023; 177:105151. DOI: 10.1016/j.ijmedinf.2023.105151. View

Zhang K, Zhang X, Ding W, Xuan N, Tian B, Huang T . National Early Warning Score Does Not Accurately Predict Mortality for Patients With Infection Outside the Intensive Care Unit: A Systematic Review and Meta-Analysis. Front Med (Lausanne). 2021; 8:704358. PMC: 8319382. DOI: 10.3389/fmed.2021.704358. View

Arabi Y, Azoulay E, Al-Dorzi H, Phua J, Salluh J, Binnie A . How the COVID-19 pandemic will change the future of critical care. Intensive Care Med. 2021; 47(3):282-291. PMC: 7898492. DOI: 10.1007/s00134-021-06352-y. View

Jehi L, Ji X, Milinovich A, Erzurum S, Merlino A, Gordon S . Development and validation of a model for individualized prediction of hospitalization risk in 4,536 patients with COVID-19. PLoS One. 2020; 15(8):e0237419. PMC: 7418996. DOI: 10.1371/journal.pone.0237419. View

Wang L, Zhang Y, Wang D, Tong X, Liu T, Zhang S . Artificial Intelligence for COVID-19: A Systematic Review. Front Med (Lausanne). 2021; 8:704256. PMC: 8514781. DOI: 10.3389/fmed.2021.704256. View

Hiscott J, Alexandridi M, Muscolini M, Tassone E, Palermo E, Soultsioti M . The global impact of the coronavirus pandemic. Cytokine Growth Factor Rev. 2020; 53:1-9. PMC: 7254014. DOI: 10.1016/j.cytogfr.2020.05.010. View

Sepandi M, Taghdir M, Alimohamadi Y, Afrashteh S, Hosamirudsari H . Factors Associated with Mortality in COVID-19 Patients: A Systematic Review and Meta-Analysis. Iran J Public Health. 2020; 49(7):1211-1221. PMC: 7548487. DOI: 10.18502/ijph.v49i7.3574. View

Rodriguez-Perez R, Bajorath J . Interpretation of machine learning models using shapley values: application to compound potency and multi-target activity predictions. J Comput Aided Mol Des. 2020; 34(10):1013-1026. PMC: 7449951. DOI: 10.1007/s10822-020-00314-0. View

Xin Y, Li H, Zhou Y, Yang Q, Mu W, Xiao H . The accuracy of artificial intelligence in predicting COVID-19 patient mortality: a systematic review and meta-analysis. BMC Med Inform Decis Mak. 2023; 23(1):155. PMC: 10410953. DOI: 10.1186/s12911-023-02256-7. View

10.

Pudjihartono N, Fadason T, Kempa-Liehr A, OSullivan J . A Review of Feature Selection Methods for Machine Learning-Based Disease Risk Prediction. Front Bioinform. 2022; 2:927312. PMC: 9580915. DOI: 10.3389/fbinf.2022.927312. View

11.

Lv C, Guo W, Yin X, Liu L, Huang X, Li S . Innovative applications of artificial intelligence during the COVID-19 pandemic. Infect Med (Beijing). 2024; 3(1):100095. PMC: 10998276. DOI: 10.1016/j.imj.2024.100095. View

12.

Linardatos P, Papastefanopoulos V, Kotsiantis S . Explainable AI: A Review of Machine Learning Interpretability Methods. Entropy (Basel). 2020; 23(1). PMC: 7824368. DOI: 10.3390/e23010018. View

13.

Haldane V, Foo C, Abdalla S, Jung A, Tan M, Wu S . Health systems resilience in managing the COVID-19 pandemic: lessons from 28 countries. Nat Med. 2021; 27(6):964-980. DOI: 10.1038/s41591-021-01381-y. View

14.

Kufel J, Bargiel-Laczek K, Kocot S, Kozlik M, Bartnikowska W, Janik M . What Is Machine Learning, Artificial Neural Networks and Deep Learning?-Examples of Practical Applications in Medicine. Diagnostics (Basel). 2023; 13(15). PMC: 10417718. DOI: 10.3390/diagnostics13152582. View

15.

Stenwig E, Salvi G, Salvo Rossi P, Skjaervold N . Comparative analysis of explainable machine learning prediction models for hospital mortality. BMC Med Res Methodol. 2022; 22(1):53. PMC: 8882271. DOI: 10.1186/s12874-022-01540-w. View

16.

Fan F, Xiong J, Li M, Wang G . On Interpretability of Artificial Neural Networks: A Survey. IEEE Trans Radiat Plasma Med Sci. 2022; 5(6):741-760. PMC: 9105427. DOI: 10.1109/trpms.2021.3066428. View

17.

Ali S, Akhlaq F, Imran A, Kastrati Z, Daudpota S, Moosa M . The enlightening role of explainable artificial intelligence in medical & healthcare domains: A systematic literature review. Comput Biol Med. 2023; 166:107555. DOI: 10.1016/j.compbiomed.2023.107555. View

18.

Bifarin O . Interpretable machine learning with tree-based shapley additive explanations: Application to metabolomics datasets for binary classification. PLoS One. 2023; 18(5):e0284315. PMC: 10159207. DOI: 10.1371/journal.pone.0284315. View

19.

Wynants L, Van Calster B, Collins G, Riley R, Heinze G, Schuit E . Prediction models for diagnosis and prognosis of covid-19: systematic review and critical appraisal. BMJ. 2020; 369:m1328. PMC: 7222643. DOI: 10.1136/bmj.m1328. View

20.

Veldhuis L, Ridderikhof M, Schinkel M, van den Bergh J, Beudel M, Dormans T . Early warning scores to assess the probability of critical illness in patients with COVID-19. Emerg Med J. 2021; 38(12):901-905. DOI: 10.1136/emermed-2020-211054. View