Portal de Periódicos da CAPES

Machine learning using the extreme gradient boosting (XGBoost) algorithm predicts 5-day delta of SOFA score at ICU admission in COVID-19 patients

2021; Elsevier BV; Volume: 1; Issue: 2 Linguagem: Inglês

10.1016/j.jointm.2021.09.002

2097-0250

Jonathan Montomoli, Luca Romeo, Sara Moccia, Michele Bernardini, Lucia Migliorelli, Daniele Berardini, Abele Donati, Andrea Carsetti, Maria Grazia Bocci, Pedro David Wendel‐Garcia, Thierry Fumeaux, Philippe Guerci, Reto Schüpbach, Can İnce, Emanuele Frontoni, Matthias P. Hilty, Mario Alfaro-Farias, Gerardo Vizmanos-Lamotte, Thomas Tschoellitsch, Jens Meier, Hernán Aguirre-Bermeo, Janina Apolo, Alberto Martínez, Geoffrey Jurkolow, Gauthier Delahaye, Emmanuel Novy, Marie-Reine Losser, Tobias Wengenmayer, Jonathan Rilinger, Dawid L. Staudacher, Sascha David, Tobias Welte, Klaus Stahl, “Agios Pavlos”, Theodoros Aslanidis, Anita Korsós, Barna Babik, Reza Nikandish, Emanuele Rezoagli, Matteo Giacomini, Alice Nova, Alberto Fogagnolo, Savino Spadaro, Roberto Ceriani, Martina Murrone, Maddalena Alessandra Wu, Chiara Cogliati, Riccardo Colombo, E Catena, F Turrini, Maria Sole Simonini, Silvia Fabbri, Antonella Potalivo, Francesca Facondini, Gianfilippo Gangitano, Tiziana Perin, Maria Grazia Bocci, Massimo Antonelli, Diederik Gommers, Raquel Rodríguez-García, Jorge Gámez-Zapata, Xiana Taboada-Fraga, Pedro Castro, Adrián Téllez, Arantxa Lander-Azcona, Jesús Escós-Orta, María Cruz Martín-Delgado, Angela Algaba-Calderon, Diego Franch-Llasat, Ferran Roche‐Campo, Herminia Lozano-Gómez, Begoña Zalba-Etayo, Marc Michot, Alexander Klarer, Rolf Ensner, Peter Schott, Severin Urech, Núria Zellweger, Lukas Merki, Adriana Lambert, Marcus Laube, Marie M. Jeitziner, Béatrice Jenni‐Moser, Jan Wiegand, Bernd Yuen, Barbara Lienhardt-Nobbe, Andrea Westphalen, Petra Salomon, Iris Drvaric, Frank Hillgaertner, Marianne Sieber, Alexander Dullenkopf, Lina Petersen, Ivan Chau, Hatem Ksouri, Govind Sridharan, Sara Cereghetti, Filippo Boroli, Jérôme Pugin, Serge Grazioli, Peter C. Rimensberger, Christian Bürkle, Julien Marrel, Mirko Brenni, Isabelle Fleisch, Jerôme Lavanchy, Marie‐Hélène Perez, Anne‐Sylvie Ramelet, Anja Baltussen Weber, Peter Gerecke, Andreas Christ, Samuele Ceruti, Andrea Glotta, Katharina Marquardt, Karim Shaikh, Tobias Hübner, Thomas A. Neff, Hermann Redecker, Mallory Moret‐Bochatay, FriederikeMeyer zu Bentrup, Michael Studhalter, Michael Stephan, Jan Brem, Nadine Gehring, Daniela Selz, Didier Naon, Gian‐Reto Kleger, Urs Pietsch, Miodrag Filipovic, Anette Ristic, Michael Sepulcri, Antje Heise, Marilene Franchitti Laurent, Jean-Christophe Laurent, Pedro David Wendel‐Garcia, Reto A. Schuepbach, Dorothea M. Heuberger, Philipp Bühler, Silvio D. Brugger, Patricia Fodor, Pascal Locher, Giovanni Camen, Tomislav Gaspert, Marija Jović, Christoph Haberthuer, Roger F. Lussman, Elif Çolak,

Intensive Care Unit Cognitive Disorders

Accurate risk stratification of critically ill patients with coronavirus disease 2019 (COVID-19) is essential for optimizing resource allocation, delivering targeted interventions, and maximizing patient survival probability. Machine learning (ML) techniques are attracting increased interest for the development of prediction models as they excel in the analysis of complex signals in data-rich environments such as critical care.We retrieved data on patients with COVID-19 admitted to an intensive care unit (ICU) between March and October 2020 from the RIsk Stratification in COVID-19 patients in the Intensive Care Unit (RISC-19-ICU) registry. We applied the Extreme Gradient Boosting (XGBoost) algorithm to the data to predict as a binary outcome the increase or decrease in patients' Sequential Organ Failure Assessment (SOFA) score on day 5 after ICU admission. The model was iteratively cross-validated in different subsets of the study cohort.The final study population consisted of 675 patients. The XGBoost model correctly predicted a decrease in SOFA score in 320/385 (83%) critically ill COVID-19 patients, and an increase in the score in 210/290 (72%) patients. The area under the mean receiver operating characteristic curve for XGBoost was significantly higher than that for the logistic regression model (0.86 vs. 0.69, P < 0.01 [paired t-test with 95% confidence interval]).The XGBoost model predicted the change in SOFA score in critically ill COVID-19 patients admitted to the ICU and can guide clinical decision support systems (CDSSs) aimed at optimizing available resources.