Determination of predictors of an unfavorable outcome in the subacute period of SARS-CoV-2 infection using machine learning methods

Background. Pathological changes in systems and organs after COVID-19 can lead to delayed death. One of the most influenced target systems of post-COVID changes is the cardiovascular system.

Aim: To identify, using machine learning methods, indicators that have predictive value in determining the adverse outcome of subacute COVID-19.

Material and Methods. The study included 212 people admitted after previous severe COVID-19. Retrospectively, the patients were divided into 2 groups: 140 patients discharged from the hospital after improvement in their state and 72 patients died during hospitalization. All patients underwent general clinical, biochemical analyses, assessment of blood coagulation system. The following machine learning methods were used for data analysis: support vector machine, random forest, stochastic gradient boosting. Validation of the obtained models was carried out by the method of 10-fold cross-validation in conjunction with ROC–AUC analysis (Receiver Operation Characteristics – Area Under Curve).

Results. In the created models, the predictors of mortality were urea and body temperature for the random forest and stochastic gradient boosting methods, erythrocyte, eosinophil and monocyte counts, and INR (International Normalized Ratio) level for the support vector machine.

Conclusion. In our study, two predictive models created using machine learning methods random forest and stochastic gradient boosting showed that changes in urea level and body temperature had predictive value. The support vector machine revealed other predictors, namely the number of erythrocytes, eosinophils and monocytes, INR. We used the voting method, on the basis of which the urea level and body temperature were established as informative signs. The random forest and stochastic gradient boosting methods showed similar results, we did not take into account the data obtained using the support vector machine. This approach of choosing a predictive model by voting is often used when evaluating data using artificial intelligence methods. It is possible that an increase in urea levels was a trigger leading to endotheliitis and subsequent myocardial infarction, before acute renal failure developed.

1. Nalbandian A., Sehgal K., Gupta A., Madhavan M.V., McGroder C., Stevens J.S. et al. Post-acute COVID-19 syndrome. Nat. Med. 2021;27(4):601–615. https://doi.org/10.1038/s41591-021-01283-z

2. Shah W., Hillman T., Playford E.D., Hishmeh L. Managing the long term effects of COVID-19: Summary of NICE, SIGN, and RCGP rapid guideline. Brit. Med. J. 2021;372:136–139. https://doi.org/10.1136/bmj.n136

3. Maestre-Muñiz M.M., Arias Á., Mata-Vázquez E., Martín-Toledano M., López-Larramona G., Ruiz-Chicote A.M. et al. Long-term outcomes of patients with coronavirus disease 2019 at one year after hospital discharge. J. Clin. Med. 2021;10(13):2945–2949. https://doi.org/10.12775/JEHS.2022.12.04.006

4. Xie Y., Xu E., Bowe B., Al-Aly Z. Long-term cardiovascular outcomes of COVID-19. Nat. Med. 2022;28(3):583–590. https://doi.org/10.1038/s41591-022-01689-3

5. Puntmann V.O., Carerj M.L., Wieters I. Outcomes of cardiovascular magnetic resonance imaging in patients recently recovered from coronavirus disease 2019 (COVID-19). JAMA Cardiol. 2020;5:1265–1273. https://doi.org/10.1001/jamacardio.2020.3557

6. Stokes E., Zambrano L., Anderson K., Marder E., Raz K., El Burai F. et al. Coronavirus disease 2019 case surveillance – United States, January 22-May 30, 2020. MMWR Morb. Mortal. Wkly Rep. 2020;69:759. https://doi.org/10.15585/mmwr.mm6924e2

7. John K.J., Mishra A.K., Ramasamy C., George A.A., Selvaraj V., Lal A. Heart failure in COVID-19 patients: critical care experience. World J. Virol. 2022;11:1–19. https://doi.org/10.5501/wjv.v11.i1.1

8. Mehandru S., Merad M. Pathological sequelae of long-haul COVID. Nat. Immunol. 2022;23(2):94–202. https://doi.org/10.1038/s41590-021-01104-y

9. Ji M., Yuan L., Shen W., Lv J., Li Y., Li M. et al. Characteristics of disease progress in patients with coronavirus disease 2019 in Wuhan, China. Epidemiology and infection. 2020;148:94–97. https://doi.org/10.1017/S0950268820000977

10. Collett D. Modelling survival data in medical research (4th ed.). New York: Chapman and Hall/CRC. 2023:556. https://doi.org/10.1201/9781003282525

11. Vrazhnov D., Mankova A., Stupak E., Kistenev Y., Shkurinov A., Cherkasova O. Discovering glioma tissue through its biomarkers' detection in blood by raman spectroscopy and machine learning. Pharmaceutics. 2023;15(1):203–215. https://doi.org/10.3390/pharmaceutics15010203

12. Gong J., Dong H., Xia Q.S., Huang Z.Y., Wang D.K., Zhao Y. et al. Correlation analysis between disease severity and inflammation-related parameters in patients with COVID-19: a retrospective study. BMC Infect. Dis. 2020;20(1):963. https://doi.org/10.1186/s12879-020-05681-5

13. Liao D., Zhou F., Luo L., Xu M., Wang H., Xia J. et al. Haematological characteristics and risk factors in the classification and prognosis evaluation of COVID-19: a retrospective cohort study. Lancet Haematol. 2020;7(9):e671–e678. https://doi.org/10.1016/S2352-3026(20)30217-9

14. Bloom P.P., Meyerowitz E.A., Reinus Z., Daidone M., Gustafson J., Kim A.Y. Liver Biochemistries in Hospitalized Patients With COVID-19. Hepatology. 2021;73(3):890–900. https://doi.org/10.1002/hep.31326

15. Bashir S., Almazroi A., Ashfaq A., Almazroi A., Khan F. A knowledge-based clinical decision support system utilizing an intelligent ensemble voting scheme for improved cardiovascular disease prediction. IEEE Access. 2021;9:130805–130822. https://doi.org/10.1109/ACCESS.2021.3110604

16. Fathima M.D., Samuel S., Natchadalingam R., Kaveri V. Majority voting ensembled feature selection and customized deep neural network for the enhanced clinical decision support system. International Journal of Computers and Applications. 2022;44(10):991–1001. https://doi.org/10.1080/1206212X.2022.2069643

17. Tharakan S., Nomoto K., Miyashita S., Ishikawa K. Body temperature correlates with mortality in COVID-19 patients. Crit. Care. 2020;24:298. https://doi.org/10.1186/s13054-020-03045-8

18. Lau W.L., Vaziri N.D. Urea, a true uremic toxin: the empire strikes back. Clin. Sci. (Lond). 2017;131(1):3–12. https://doi.org/10.1042/CS20160203

19. d'Apolito M,. Colia A.L., Manca E., Pettoello-Mantovani M., Sacco M., Maffione A.B., Brownlee M., Giardino I. Urea memory: transient cell exposure to urea causes persistent mitochondrial ROS production and endothelial dysfunction. Toxins (Basel). 2018;10(10):410. https://doi.org/10.3390/toxins10100410

20. Colombo G., Altomare A., Astori E., Landoni L., Garavaglia M.L., Rossi R. et al. Effects of Physiological and Pathological Urea Concentrations on Human Microvascular Endothelial Cells. Int. J. Mol. Sci. 2022;24(1):691. https://doi.org/10.3390/ijms24010691

21. Ertuğlu L.A., Kanbay A., Afşar B., Elsürer Afşar R., Kanbay M. COVID-19 and acute kidney injury. Tuberk. Toraks. 2020;68(4):407–418. https://doi.org/10.5578/tt.70010