A value of respiratory muscle strength indicators to determine severity of chronic obstructive pulmonary disease using artificial neural networks

The aim of the study was to analyze a diagnostic value of respiratory muscle (RM) strength indicators to assess severity of chronic obstructive pulmonary disease (COPD) using machine learning methods and artificial neural networks (ANN). Methods. One hundred and fifteen males with acute exacerbation  of COPD  were involved in the study. RM strength indicators (MEP,  MIP,  and SNIP), demographic  parameters,  spirometry,  blood gases, dyspnea with mMRC and CAT scales were measured. Statistical analysis was performed using Mann-Whitney’s, Fisher’s and Tukey’s tests and correlation  analysis. RM strength model was performed  using linear and nonlinear  regression analysis. COPD  stratification  model was performed using ANN. Results. RM strength models in healthy males and COPD  patients allowed estimation  the impact of different factors on the RM functional status. Comparison  of COPD  stratification  for severity using the mathematical model or expert diagnosis showed that combination of FEV1 with other indicators could increase the accuracy of ANN model. MIP,  the total body mass, partial CO2  tension in the arterial blood and serum fibrinogen concentration were the most valuable indicators.  Moreover,  MIP  was considered  as the universal predictor  increasing the accuracy of all models. Conclusion. Practical application  of ANN models in telemedicine  projects is related to the improvement  of ANN architecture and development of informational services which would allow a real-time assessment of the patient's condition.

1. Kaminska M., Noel F., Petrof B.J. Optimal method for assessment of respiratory muscle strength in neuromuscular disorders using sniff nasal inspiratory pressure (SNIP). PLoS One. 2017; 12 (5): e0177723. DOI: 10.1371/journal.pone.0177723.

2. Byrd R.B., Hyatt R.E. Maximal respiratory pressures in chronic obstructive lung disease. Am. Rev. Respir. Dis. 1968; 98 (5): 848–856.

3. Rochester D.F., Braun N.M., Arora N.S. Respiratory muscle strength in chronic obstructive pulmonary disease. Am. Rev. Respir. Dis. 1979; 119 (2, Pt 2): 151–154.

4. Decramer M., Demedts M., Rochette F. et al. Maximal transrespiratory pressures in obstructive lung disease. Bull. Eur. Physiopathol. Respir. 1980; 16 (4): 479–490.

5. Morrison N.J., Richardson J., Dunn L., Pardy R.L. Respiratory muscle performance in normal elderly subjects and patients with COPD. Chest. 1989; 95 (1): 90–94.

6. Rochester D.F., Braun N.M.T. Determinants of maximal inspiratory pressure in chronic obstructive pulmonary disease. J. Crit. Care. 1986; 1 (1): 61–62. DOI: 10.1016/s0883-9441(86)80138-1.

7. Terzano C., Ceccarelli D., Conti V. et al. Maximal respiratory static pressures in patients with different stages of COPD severity. Respir. Res. 2008; 9 (1): 8. DOI: 10.1186/1465-9921-9-8.

8. Hafez M.R., Elsheikh R.M. Assessment of the Respiratory Muscles Function in Chronic Obstructive Pulmonary Disease patients. Egypt. J. Hosp. Med. 2012; 49: 661–671.

9. Formiga M.F., Campos M.A., Cahalin L.P. et al. Sustained maximal inspiratory pressure is significantly related to mortality risk in COPD. Eur. Respir. J. 2017; 50 (Suppl. 61): PA2007. DOI: 10.1183/1393003.congress-2017.PA2007.

10. Khalil M., Wagih K., Mahmoud O. Evaluation of maximum inspiratory and expiratory pressure in patients with chronic obstructive pulmonary disease. Egypt. J. Chest Dis. Tuberc. 2014; 63 (2): 329–335. DOI: 10.1016/j.ejcdt.2014.01.010.

11. Kim N.S., Seo J.H., Ko M.H. et al. Respiratory muscle strength in patients with chronic obstructive pulmonary disease. Ann. Rehabil. Med. 2017; 41 (4): 659–666. DOI: 10.5535/arm.2017.41.4.659.

12. van der Palen J., Rea T.D., Manolio T.A. et al. Respiratory muscle strength and the risk of incident cardiovascular events. Thorax. 2004; 59 (12): 1063–1067. DOI: 10.1136/thx.2004.021915.

13. Volaklis K.A., Halle M., Meisinger C. Muscular strength as a strong predictor of mortality: a narrative review. Eur. J. Intern. Med. 2015; 26 (5): 303–310. DOI: 10.1016/j.ejim.2015.04.013.

14. Gel'tser B.I., Kotel'nikov V.N., Shakhgel'dyan K.I., Kurpatov I.G. [Results of modeling of predicted values for respiratory muscle strength using artif icial neuronal networks]. Rossiyskiy fiziologicheskiy zhurnal im. I.M.Sechenova. 2018; 104 (9): 1065–1074. DOI: 10.7868/S0869813918090058 (in Russian).

15. Chuchalin A.G., Avdeev S.N., Aisanov Z.R. et al. Federal Clinical Guidelines on Diagnosis and Management of Chronic Obstructive Pulmonary Disease. Pul'monologiya. 2014; (3): 15–54. DOI: 10.18093/0869-0189-2014-0-3-15-54 (in Russian).

16. Filipenko P.S., Shutko I.S. [An assessment of toxication severity in patients with cavitating pneumonia and acute lung abscess]. Meditsinskiy vestnik Severnogo Kavkaza. 2010; 20 (4): 14–16. Available at: https://cyberleninka.ru/article/n/otsenka-stepeni-intoksikatsii-pri-destruktivnoy-pnevmonii-i-ostrom-abstsesse-legkogo/viewer (in Russian).

17. Shchankin A.A. [An association between body composition and physiological functions]. Moscow – Berlin: Direkt-Media Pablishing; 2015 (in Russian).

18. Hyndman R.J., Athanasopoulos G. Forecasting: principles and practice. Sydney: OTexts; 2013.

19. Le Cessie S., van Houwelingen J.C. Ridge estimators in logistic regression. J. Royal Stat. Soc. 1992; 41 (1): 191–201. DOI: 10.2307/2347628.

20. Vorontsov K.V. [A combinatory approach to quality assessment of machine learning]. In.: Lupanov O.B., ed. Matematicheskie voprosy kibernetiki. Moscow: Fizmatlit; 2004. Vol. 13: 5–36 (in Russian).

21. Evans J.A., Whitelaw W.A. The assessment of maximal respiratory mouth pressures in adults. Respir. Care. 2009; 54 (10): 1348–1359.

22. Gel'tser B.I., Kurpatov I.G., Kotel'nikov V.N. Parameters of respiratory muscle strength in healthy persons: age-related, gender and constitutional features. Rossiyskiy fiziologicheskiy zhurnal im. I.M.Sechenova. 2017; 103 (12): 1425–1433 (in Russian).

23. Abrosimov V.N., Ponomareva L.B., Osychnaya N.A. [Respiratory muscle dysfunction and its diagnosis in patients with chronic obstructive pulmonary disease]. Klinicheskaya gerontologiya. 2008; 14 (6): 38–43. Available at: https://cyberleninka.ru/article/n/respiratornaya-myshechnaya-disfunktsiya-i-ee-diagnostika-u-bolnyh-s-hronicheskoy-obstruktivnoy-boleznyu-legkih/viewer (in Russian).

24. Kabakov R.I. [Data analysis and visualization using the R language]. Translation from English – Volkova P.D. Moscow: DMK Press; 2014 (in Russian).

25. Mun S.A., Glushov A.N., Shternis T.A. et al. [Regression analysis in medical and biological studies]. Kemerovo: KemGMA; 2012 (in Russian).

26. Aleksandrova N.P., Breslav I.S. [Human respiratory muscles: three levels of the control]. Fiziologiya cheloveka. 2009; 35 (2): 103–111 (in Russian).

27. Artuğ N.T., Goker I., Bolat B. et al. New features for scanned bioelectrical activity of motor unit in health and disease. Biomed. Signal. Proces. Control. 2018; 41: 109–128. DOI: 10.1016/j.bspc.2017.11.011.

28. Fernandes M., Cukier A., Ambrosino N. et al. Respiratory pattern, thoracoabdominal motion and ventilation in chronic airway obstruction. Monaldi Arch. Chest Dis. 2007; 67(4): 209–216.

29. Maltais F., Decramer M., Casaburi R. et al. An Official American Thoracic Society/European Respiratory Society statement: update on limb muscle dysfunction in chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med. 2014; 189 (9): е15–62. DOI: 10.1164/rccm.201402-0373st.

30. Tudorache V., Oancea C., Mlădinescu O.F. Clinical relevance of maximal inspiratory pressure: determination in COPD exacerbation. Intern. J. Chron. Obstruct. Pulmon. Dis. 2010; 5: 119–123 DOI: 10.2147/copd.s9194.