Generic placeholder image

Current Pharmaceutical Biotechnology

Author(s): Peng Yan, Siwan Huang, Ye Li, Tiange Chen, Xiang Li, Yuan Zhang, Huan Wu, Jianqiao Xu, Guotong Xie, Lixin Xie* and Guoxin Mo*

DOI: 10.2174/1389201024666230220103755

DownloadDownload PDF Flyer Cite As
A New Risk Model based on the Machine Learning Approach for Prediction of Mortality in the Respiratory Intensive Care Unit

Page: [1673 - 1681] Pages: 9

  • * (Excluding Mailing and Handling)

Abstract

Background: Intensive care unit (ICU) resources are inadequate for the large population in China, so it is essential for physicians to evaluate the condition of patients at admission. In this study, our objective was to construct a machine-learning risk prediction model for mortality in respiratory intensive care units (RICUs).

Methods: This study involved 817 patients who made 1,063 visits and who were admitted to the RICU from 2012 to 2017. Potential predictors such as demographic information, laboratory results, vital signs and clinical characteristics were considered. We constructed eXtreme Gradient Boosting (XGBoost) models and compared performances with random forest models, logistic regression models and clinical scores such as Acute Physiology and Chronic Health Evaluation II (APACHE II) and the sequential organ failure assessment (SOFA) system. The model was externally validated using data from Medical Information Mart for Intensive Care (MIMIC-III) database. A web-based calculator was developed for practical use.

Results: Among the 1,063 visits, the RICU mortality rate was 13.5%. The XGBoost model achieved the best performance with the area under the receiver operating characteristics curve (AUROC) of 0.860 (95% confidence interval (CI): 0.808 - 0.909) in the test set, which was significantly greater than APACHE II (0.749, 95% CI: 0.674 - 0.820; P = 0.015) and SOFA (0.751, 95% CI: 0.669 - 0.818; P = 0.018). The Hosmer-Lemeshow test indicated a good calibration of our predictive model in the test set with a P-value of 0.176. In the external validation dataset, the AUROC of XGBoost model was 0.779 (95% CI: 0.714 - 0.813). The final model contained variables that were previously known to be associated with mortality, but it also included some features absent from the clinical scores. The mean N-terminal pro-B-type natriuretic peptide (NTproBNP) of survivors was significantly lower than that of the non-survival group (2066.43 pg/mL vs. 8232.81 pg/mL; P < 0.001).

Conclusions: Our results showed that the XGBoost model could be a suitable model for predicting RICU mortality with easy-to-collect variables at admission and help intensivists improve clinical decision-making for RICU patients. We found that higher NT-proBNP can be a good indicator of poor prognosis.

Graphical Abstract

[1]
Halpern, N.A.; Pastores, S.M. Critical care medicine in the United States 2000-2005: An analysis of bed numbers, occupancy rates, payer mix, and costs. Crit. Care Med., 2010, 38(1), 65-71.
[http://dx.doi.org/10.1097/CCM.0b013e3181b090d0] [PMID: 19730257]
[2]
Du, B.; Xi, X.; Chen, D.; Peng, J. Clinical review: Critical care medicine in mainland China. Crit. Care, 2010, 14(1), 206.
[http://dx.doi.org/10.1186/cc8222] [PMID: 20236446]
[3]
Knaus, W.A.; Draper, E.A.; Wagner, D.P.; Zimmerman, J.E. APACHE II: A severity of disease classification system. Crit. Care Med., 1985, 13(10), 818-829.
[http://dx.doi.org/10.1097/00003246-198510000-00009] [PMID: 3928249]
[4]
Diaz-Flores, E.; Meyer, T.; Giorkallos, A. Evolution of artificial intelligence-powered technologies in biomedical research and healthcare. Adv. Biochem. Eng. Biotechnol., 2022, 182, 23-60.
[http://dx.doi.org/10.1007/10_2021_189] [PMID: 35262750]
[5]
Aung, Y.Y.M.; Wong, D.C.S.; Ting, D.S.W. The promise of artificial intelligence: A review of the opportunities and challenges of artificial intelligence in healthcare. Br. Med. Bull., 2021, 139(1), 4-15.
[http://dx.doi.org/10.1093/bmb/ldab016] [PMID: 34405854]
[6]
Vincent, J.L.; de Mendonça, A.; Cantraine, F.; Moreno, R.; Takala, J.; Suter, P.M.; Sprung, C.L.; Colardyn, F.; Blecher, S. Use of the SOFA score to assess the incidence of organ dysfunction/failure in intensive care units. Crit. Care Med., 1998, 26(11), 1793-1800.
[http://dx.doi.org/10.1097/00003246-199811000-00016] [PMID: 9824069]
[7]
Zheng, T.; Xie, W.; Xu, L.; He, X.; Zhang, Y.; You, M.; Yang, G.; Chen, Y. A machine learning-based framework to identify type 2 diabetes through electronic health records. Int. J. Med. Inform., 2017, 97, 120-127.
[http://dx.doi.org/10.1016/j.ijmedinf.2016.09.014] [PMID: 27919371]
[8]
Fernández-Gonzalo, S.; Navarra-Ventura, G.; Bacardit, N.; Gomà Fernández, G.; de Haro, C.; Subirà, C.; López-Aguilar, J.; Magrans, R.; Sarlabous, L.; Aquino Esperanza, J.; Jodar, M.; Rué, M.; Ochagavía, A.; Palao, D.J.; Fernández, R.; Blanch, L. Cognitive phenotypes 1 month after ICU discharge in mechanically ventilated patients: A prospective observational cohort study. Crit. Care, 2020, 24(1), 618.
[http://dx.doi.org/10.1186/s13054-020-03334-2] [PMID: 33087171]
[9]
Johnson, A.E.W.; Pollard, T.J.; Shen, L.; Lehman, L.H.; Feng, M.; Ghassemi, M.; Moody, B.; Szolovits, P.; Anthony Celi, L.; Mark, R.G. MIMIC-III, a freely accessible critical care database. Sci. Data, 2016, 3(1), 160035.
[http://dx.doi.org/10.1038/sdata.2016.35] [PMID: 27219127]
[10]
Chen, T.; Guestrin, C. XGBoost: A scalable tree boosting system. 22nd SIGKDD conference on knowledge discovery and data mining, 2016, 785-794.
[11]
Wang, L.; Wang, X.; Chen, A.; Jin, X.; Che, H. Prediction of type 2 diabetes risk and its effect evaluation based on the XGBoost model. Health care, 2020, 8(3), 247.
[http://dx.doi.org/10.3390/healthcare8030247] [PMID: 32751894]
[12]
Huang, J.C.; Tsai, Y.C.; Wu, P.Y.; Lien, Y.H.; Chien, C.Y.; Kuo, C.F.; Hung, J.F.; Chen, S.C.; Kuo, C.H. Predictive modeling of blood pressure during hemodialysis: A comparison of linear model, random forest, support vector regression, XGBoost, LASSO regression and ensemble method. Comput. Methods Programs Biomed., 2020, 195, 105536.
[http://dx.doi.org/10.1016/j.cmpb.2020.105536] [PMID: 32485511]
[13]
Al Daoud, E. Comparison between XGBoost, LightGBM and CatBoost using a home credit dataset. IJCIT, 2019, 13(1), 6-10.
[14]
Breiman, L. Random forests. Mach. Learn., 2001, 45(1), 5-32.
[http://dx.doi.org/10.1023/A:1010933404324]
[15]
Tibshirani, R. Regression shrinkage and selection via the lasso. J. R. Stat. Soc. B, 1996, 58(1), 267-288.
[http://dx.doi.org/10.1111/j.2517-6161.1996.tb02080.x]
[16]
Lundberg, S.; Lee, S. A unified approach to interpreting model predictions. Adv. Neural Inf. Process. Syst., 2017, 4765-4774.
[17]
Mangalathu, S.; Hwang, S.H.; Jeon, J.S. Failure mode and effects analysis of RC members based on machine-learning-based shapley additive explanations (SHAP) approach. Eng. Struct., 2020, 219, 110927.
[http://dx.doi.org/10.1016/j.engstruct.2020.110927]
[18]
DeLong, E.R.; DeLong, D.M.; Clarke-Pearson, D.L. Comparing the areas under two or more correlated receiver operating characteristic curves: A nonparametric approach. Biometrics, 1988, 44(3), 837-845.
[http://dx.doi.org/10.2307/2531595] [PMID: 3203132]
[19]
Lemeshow, S.; Hosmer, D.W. A review of goodness of fit statistics for use in the development of logistic regression models. Am. J. Epidemiol., 1982, 115(1), 92-106.
[http://dx.doi.org/10.1093/oxfordjournals.aje.a113284] [PMID: 7055134]
[20]
Liang, W.; Liang, H.; Ou, L.; Chen, B.; Chen, A.; Li, C.; Li, Y.; Guan, W.; Sang, L.; Lu, J.; Xu, Y.; Chen, G.; Guo, H.; Guo, J.; Chen, Z.; Zhao, Y.; Li, S.; Zhang, N.; Zhong, N.; He, J. Development and validation of a clinical risk score to predict the occurrence of critical illness in hospitalized patients with COVID-19. JAMA Intern. Med., 2020, 180(8), 1081-1089.
[http://dx.doi.org/10.1001/jamainternmed.2020.2033] [PMID: 32396163]
[21]
Pan, P.; Li, Y.; Xiao, Y.; Han, B.; Su, L.; Su, M.; Li, Y.; Zhang, S.; Jiang, D.; Chen, X.; Zhou, F.; Ma, L.; Bao, P.; Xie, L. Prognostic assessment of COVID-19 in the intensive care unit by machine learning methods: Model development and validation. J. Med. Internet Res., 2020, 22(11), e23128.
[http://dx.doi.org/10.2196/23128] [PMID: 33035175]
[22]
Quiroz-Juárez, M.A.; Torres-Gómez, A.; Hoyo-Ulloa, I.; León-Montiel, R.J. U’Ren, A.B. Identification of high-risk COVID-19 patients using machine learning. PLoS One, 2021, 16(9), e0257234.
[http://dx.doi.org/10.1371/journal.pone.0257234] [PMID: 34543294]
[23]
Ghassemi, M.; Naumann, T.; Doshi-Velez, F.; Brimmer, N.; Joshi, R.; Rumshisky, A.; Szolovits, P. Unfolding physiological state: Mortality modelling in intensive care units. KDD, 2014, 2014, 75-84.
[http://dx.doi.org/10.1145/2623330.2623742] [PMID: 25289175]
[24]
Meyer, A.; Zverinski, D.; Pfahringer, B.; Kempfert, J.; Kuehne, T.; Sündermann, S.H.; Stamm, C.; Hofmann, T.; Falk, V.; Eickhoff, C. Machine learning for real-time prediction of complications in critical care: A retrospective study. Lancet Respir. Med., 2018, 6(12), 905-914.
[http://dx.doi.org/10.1016/S2213-2600(18)30300-X] [PMID: 30274956]
[25]
Hug, C.W.; Szolovits, P. ICU acuity: Real-time models versus daily models. AMIA Annu. Symp. Proc., 2009, 2009, 260-264.
[PMID: 20351861]
[26]
de Lemos, J.A.; Morrow, D.A.; Bentley, J.H.; Omland, T.; Sabatine, M.S.; McCabe, C.H.; Hall, C.; Cannon, C.P.; Braunwald, E. The prognostic value of B-type natriuretic peptide in patients with acute coronary syndromes. N. Engl. J. Med., 2001, 345(14), 1014-1021.
[http://dx.doi.org/10.1056/NEJMoa011053] [PMID: 11586953]
[27]
Meyer, B.; Huelsmann, M.; Wexberg, P.; Karth, G.D.; Berger, R.; Moertl, D.; Szekeres, T.; Pacher, R.; Heinz, G. N-terminal pro-B-type natriuretic peptide is an independent predictor of outcome in an unselected cohort of critically ill patients. Crit. Care Med., 2007, 35(10), 2268-2273.
[http://dx.doi.org/10.1097/01.CCM.0000284509.23439.5B] [PMID: 17717488]
[28]
Dhondup, T.; Tien, J.C.C.; Marquez, A.; Kennedy, C.C.; Gajic, O.; Kashani, K.B. Association of negative fluid balance during the de-escalation phase of sepsis management with mortality: A cohort study. J. Crit. Care, 2020, 55, 16-21.
[http://dx.doi.org/10.1016/j.jcrc.2019.09.025] [PMID: 31670149]
[29]
Sadaka, F.; Juarez, M.; Naydenov, S.; O’Brien, J. Fluid resuscitation in septic shock: The effect of increasing fluid balance on mortality. J. Intensive Care Med., 2014, 29(4), 213-217.
[http://dx.doi.org/10.1177/0885066613478899] [PMID: 23753235]
[30]
Laffey, J.G.; Bellani, G.; Pham, T.; Fan, E.; Madotto, F.; Bajwa, E.K.; Brochard, L.; Clarkson, K.; Esteban, A.; Gattinoni, L.; van Haren, F.; Heunks, L.M.; Kurahashi, K.; Laake, J.H.; Larsson, A.; McAuley, D.F.; McNamee, L.; Nin, N.; Qiu, H.; Ranieri, M.; Rubenfeld, G.D.; Thompson, B.T.; Wrigge, H.; Slutsky, A.S.; Pesenti, A. Potentially modifiable factors contributing to outcome from acute respiratory distress syndrome: The LUNG SAFE study. Intensive Care Med., 2016, 42(12), 1865-1876.
[http://dx.doi.org/10.1007/s00134-016-4571-5] [PMID: 27757516]
[31]
Bagshaw, S.M.; Webb, S.A.R.; Delaney, A.; George, C.; Pilcher, D.; Hart, G.K.; Bellomo, R. Very old patients admitted to intensive care in Australia and New Zealand: A multi-centre cohort analysis. Crit. Care, 2009, 13(2), R45.
[http://dx.doi.org/10.1186/cc7768] [PMID: 19335921]
[32]
Grasselli, G.; Greco, M.; Zanella, A.; Albano, G.; Antonelli, M.; Bellani, G.; Bonanomi, E.; Cabrini, L.; Carlesso, E.; Castelli, G.; Cattaneo, S.; Cereda, D.; Colombo, S.; Coluccello, A.; Crescini, G.; Forastieri Molinari, A.; Foti, G.; Fumagalli, R.; Iotti, G.A.; Langer, T.; Latronico, N.; Lorini, F.L.; Mojoli, F.; Natalini, G.; Pessina, C.M.; Ranieri, V.M.; Rech, R.; Scudeller, L.; Rosano, A.; Storti, E.; Thompson, B.T.; Tirani, M.; Villani, P.G.; Pesenti, A.; Cecconi, M.; Agosteo, E.; Albano, G.; Albertin, A.; Alborghetti, A.; Aldegheri, G.; Antonini, B.; Barbara, E.; Bardelloni, G.; Basilico, S.; Belgiorno, N.; Bellani, G.; Beretta, E.; Berselli, A.; Bianciardi, L.; Bonanomi, E.; Bonazzi, S.; Borelli, M.; Bottino, N.; Bronzini, N.; Brusatori, S.; Cabrini, L.; Capra, C.; Carnevale, L.; Castelli, G.; Catena, E.; Cattaneo, S.; Cecconi, M.; Celotti, S.; Cerutti, S.; Chiumello, D.; Cirri, S.; Citerio, G.; Colombo, S.; Coluccello, A.; Coppini, D.; Corona, A.; Cortellazzi, P.; Costantini, E.; Covello, R.D.; Crescini, G.; De Filippi, G.; Dei Poli, M.; Dughi, P.; Fieni, F.; Florio, G.; Forastieri Molinari, A.; Foti, G.; Fumagalli, R.; Galletti, M.; Gallioli, G.A.; Gay, H.; Gemma, M.; Gnesin, P.; Grasselli, G.; Greco, S.; Greco, M.; Grosso, P.; Guatteri, L.; Guzzon, D.; Iotti, G.A.; Keim, R.; Langer, T.; Latronico, N.; Lombardo, A.; Lorini, F.L.; Mamprin, F.; Marino, G.; Marino, F.; Merli, G.; Micucci, A.; Militano, C.R.; Mojoli, F.; Monti, G.; Muttini, S.; Nadalin, S.; Natalini, G.; Perazzo, P.; Perego, G.B.; Perotti, L.; Pesenti, A.; Pessina, C.M.; Petrucci, N.; Pezzi, A.; Piva, S.; Portella, G.; Protti, A.; Racagni, M.; Radrizzani, D.; Raimondi, M.; Ranucci, M.; Rech, R.; Riccio, M.; Rosano, A.; Ruggeri, P.; Sala, G.; Salvi, L.; Sebastiano, P.; Severgnini, P.; Sigurtà, D.; Stocchetti, N.; Storti, E.; Subert, M.; Tavola, M.; Todaro, S.; Torriglia, F.; Tubiolo, D.; Valsecchi, R.; Villani, P.G.; Viola, U.; Vitale, G.; Zambon, M.; Zanella, A.; Zoia, E. Risk factors associated with mortality among patients with COVID-19 in intensive care units in Lombardy, Italy. JAMA Intern. Med., 2020, 180(10), 1345-1355.
[http://dx.doi.org/10.1001/jamainternmed.2020.3539] [PMID: 32667669]
[33]
Chen, Y.; Klein, S.L.; Garibaldi, B.T.; Li, H.; Wu, C.; Osevala, N.M.; Li, T.; Margolick, J.B.; Pawelec, G.; Leng, S.X. Aging in COVID-19: Vulnerability, immunity and intervention. Ageing Res. Rev., 2021, 65, 101205.
[http://dx.doi.org/10.1016/j.arr.2020.101205] [PMID: 33137510]
[34]
Oh, T.; Lee, J.; Lee, Y.; Hwang, J.W.; Do, S.H.; Jeon, Y.T.; Song, I.A. Association between modified body mass index and 30-day and 1-year mortality after intensive care unit admission: A retrospective cohort study. J. Clin. Med., 2018, 7(4), 81.
[http://dx.doi.org/10.3390/jcm7040081] [PMID: 29652842]