Assessing the risk of developing moderate and severe pneumonia in COVID‑19 patients using machine learning methods

The work is devoted to the development of a computer method for assessing the severity of pneumonia, which is a complication of COVID‑19, based on sets of clinical and laboratory indicators using machine learning (ML) methods. The work consisted of investigating the possibility of predicting using ML the computed tomography (CT) grade of severity of pneumonia. Groups of 31 patients with moderate or severe pneumonia (CT2–CT4) and 113 patients without pneumonia or with mild pneumonia (CT0–CT1) were compared. The database included 105 clinical and laboratory parameters. To compare groups, standard nonparametric χ² tests and the Mann-Whitney test (U-test) with Bonferroni-Holm correction for multiple testing were used. Along with traditional statistical methods, an original method of data mining was also used, allowing statistically sound identification of informative intervals of indicator values. To predict severity gradations, a set of ML methods was used, which included, along with widely known methods, also an original development. The research used the Data Master Azforus software package. The study confirmed the possibility of predicting the results of CT based severity grades by clinical and laboratory indicators using machine learning methods. The forecast efficiency according to ROC AUC was about 0.9. The introduction of the model into practice will help improve the accuracy and efficiency of diagnosing severe pneumonia.

1. Bilichenko T.N. Risk factors, immunological mechanisms and biological markers of severe COVID 19 (review of studies). RMZh. Medical Review. 2021. Vol. 5, No. 5. P. 237–244. (In Russ.). DOI: http://doi.org/10.32364/2587 6821 2021 5 5 237 244

2. Kudryavtsev Yu.S., Beregov M.M., Berdalin A.B., Lelyuk V.G. Comparison of the main scales for assessing the severity of lung damage in COVID 19 based on computed tomography data and assessment of their prognostic value. Bulletin of Roentgenology and Radiology. 2021. Vol. 102, No. 5. P. 296–303. (In Russ.). https://doi.org/10.20862/0042 4676 2021 102 5 296 303

3. Hemraj K.S., Jacob M.J., Kotian V., Sachin D.K., Geetha R.G., & Veliath L.B. Correlation of CT Severity Index with Clinical Severity of COVID 19 Pneumonia and it’s Relation to Disease Outcome // Journal of Advances in Medicine and Medical Research. 2022. Vol. 34, № 12. P. 16–26. https://doi.org/10.9734/jammr/2022/v34i1231369

4. Inoue A., Takahashi H., Ibe T. et al. Comparison of semiquantitative chest CT scoring systems to estimate severity in coronavirus disease 2019 (COVID 19) pneumonia // Eur Radiol. 2022. Vol. 32, P. 3513–3524. https://doi.org/10.1007/s00330 021 08435 2

5. Almasi N.A., Shahnazari R., Abbasi M.A., Divsalar F., Bayazidi M., Sadatnaseri A. CT severity score in COVID 19 patients, assessment of performance in triage and outcome prediction: a comparative study of different methods // Egypt. J. Radiol. Nucl. Med. 2022. Vol. 53, № 1. P. 116. https://doi:10.1186/s43055 022 00781 5 Epub. 2022 May 18. PMCID: PMC9116070.

6. Szabó M., Kardos Z., Kostyál L., Tamáska P., Oláh C., Csánky E. and Szekanecz Z. The importance of chest CT severity score and lung CT patterns in risk assessment in COVID 19- associated pneumonia: a comparative study // Front. Med. 2023. Vol. 10. P. 1125530. https://doi:10.3389/fmed.2023.1125530

7. Morozov S.P., Gombolevsky V.A., Chernina V. Yu. et al. Prediction of fatal outcomes in COVID 19 based on chest computed tomography data // Tuberculosis and lung diseases. 2020. Vol. 98, No. 6. P. 7–14. (In Russ.). https://doi.org/10.21292/207 5 1230 2020 98 6 7 14

8. Melekhov A. V., Saifullin M. A., Petrovichev V. S., Agaeva A. I., Golubykh K. Yu., Nikitin I.G. Comparison of computed tomography data with outcomes, clinical and laboratory characteristics of patients with COVID 19 // Archives of Internal Medicine. 2021. Vol. 11, No. 6. P. 447–456. (In Russ.). https://doi.org/10.20514/222 6 6704 2021 11 6 447 456

9. Sharma S, Aggarwal A, Sharma RK, Patras E, Singhal A. Correlation of chest CT severity score with clinical parameters in COVID 19 pulmonary disease in a tertiary care hospital in Delhi during the pandemic period // Egypt J Radiol Nucl Med. 2022. Vol. 53, № 1. P. 166. DOI: 10.1186/s43055–022–00832-x. Epub 2022 Jul 28. PMCID: PMC9330926.

10. Wang D, Willis DR, Yih Y. The pneumonia severity index: Assessment and comparison to popular machine learning classifiers // Int J Med Inform. 2022. Jul, Vol. 163, P. 104778. DOI: 10.1016/j.ijmedinf.2022.104778. Epub 2022 Apr 26. PMID: 35487075.

11. Holm, S. A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics. 1979. Vol. 6. № 2. P. 65–70. JSTOR4615733. MR0538597 12. Senko O.V., Kuznetsova A.V., Matveev I.A., Litvinchev I.S. Chapter 8 – Search of regularities in data: optimality, validity, and interpretability, Editor(s): Pandian Vasant, Joshua Thomas, Elias Munapo, Gerhard-Wilhelm Weber, Advances of Artificial Intelligence in a Green Energy Environment, Academic Press, 2022, P. 151–171.

12. Senko O.V., Kuznetsova A.V., Voronin E.M., Kravtsova O.A., Borisova L.R., Kirilyuk I.L., Akimkin V.G. Methods of data mining in COVID 19 epidemic research. J. of Belarusian State University. Mat. Inf. 2022. Vol. 1, pp. 83–96. (In Russ.).

13. Cox D.R. The regression analysis of binary sequences (with discussion). J R Stat Soc B. 1958; 20(2):215–242. https://doi.org/10.1111/j.2517–6161.1958.tb00292.x

14. Cortes C., Vapnik V. Support-Vector Networks. Machine Learning, 20, 273–297 (1995) DOI: 10.1007/BF00994018

15. Hastie, T., Tibshirani R., Friedman J.. The Elements of Statistical Learning: Data Mining, Inference, and Prediction. 2nd ed.– Springer-Verlag, 2009. ISBN978 0 387 84857 0.

16. Friedman, J. H. (2002). Stochastic gradient boosting. Comput. Stat. Data Anal., 38 (4): 367–378.

17. Chen, T. and Guestrin, C. (2016). Xgboost: A scalable tree boosting system. In Proceedings of the 22Nd ACM SIGKDD International Conference on Knowl- edge Discovery and Data Mining, KDD ’16, pages 785–794, New York, NY, USA. ACM.

18. Kuznetsova, A.V., Kostomarova I.V., Senko O.V. Modification of the method of optimal valid partitioning for comparison of patterns related to the occurrence of ischemic stroke in two groups of patients. Pattern Recognition and Image Analysis 24 (1), 114–123

19. Guliev R.R., Senko O.V., et al. Application of optimal partitions for multiparameter data analysis in clinical trials. Mathematical biology and bioinformatics 2016. Vol. 11. No. 1. P. 46–63. (In Russ.). DOI: 10.17537/2016.11.46

20. Zweig, Mark H.; Campbell, Gregory (1993). Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. Clinical Chemistry. 39 (8): 561–577. DOI: 10.1093/clinchem/39.4.561. PMID8472349.