scholarly journals Heterogeneous effects of alveolar recruitment in acute respiratory distress syndrome: a machine learning reanalysis of the Alveolar Recruitment for Acute Respiratory Distress Syndrome Trial

2019 ◽  
Vol 123 (1) ◽  
pp. 88-95 ◽  
Author(s):  
Fernando G. Zampieri ◽  
Eduardo L. Costa ◽  
Theodore J. Iwashyna ◽  
Carlos R.R. Carvalho ◽  
Lucas P. Damiani ◽  
...  
Keyword(s):  
Machine Learning ◽  
Acute Respiratory Distress Syndrome ◽  
Respiratory Distress Syndrome ◽  
Respiratory Distress ◽  
Distress Syndrome ◽  
Alveolar Recruitment ◽  
Heterogeneous Effects

scholarly journals Predictive model for acute respiratory distress syndrome events in ICU patients in China using machine learning algorithms: a secondary analysis of a cohort study

2019 ◽  
Vol 17 (1) ◽  
Author(s):  
Xian-Fei Ding ◽  
Jin-Bo Li ◽  
Huo-Yan Liang ◽  
Zong-Yu Wang ◽  
Ting-Ting Jiao ◽  
...  
Keyword(s):  
Machine Learning ◽  
Acute Respiratory Distress Syndrome ◽  
Cohort Study ◽  
Respiratory Distress Syndrome ◽  
Respiratory Distress ◽  
Distress Syndrome ◽  
External Validation ◽  
Secondary Analysis ◽  
Machine Learning Algorithms ◽  
Chinese Patients

Abstract Background To develop a machine learning model for predicting acute respiratory distress syndrome (ARDS) events through commonly available parameters, including baseline characteristics and clinical and laboratory parameters. Methods A secondary analysis of a multi-centre prospective observational cohort study from five hospitals in Beijing, China, was conducted from January 1, 2011, to August 31, 2014. A total of 296 patients at risk for developing ARDS admitted to medical intensive care units (ICUs) were included. We applied a random forest approach to identify the best set of predictors out of 42 variables measured on day 1 of admission. Results All patients were randomly divided into training (80%) and testing (20%) sets. Additionally, these patients were followed daily and assessed according to the Berlin definition. The model obtained an average area under the receiver operating characteristic (ROC) curve (AUC) of 0.82 and yielded a predictive accuracy of 83%. For the first time, four new biomarkers were included in the model: decreased minimum haematocrit, glucose, and sodium and increased minimum white blood cell (WBC) count. Conclusions This newly established machine learning-based model shows good predictive ability in Chinese patients with ARDS. External validation studies are necessary to confirm the generalisability of our approach across populations and treatment practices.

scholarly journals Supervised machine learning for the early prediction of acute respiratory distress syndrome (ARDS)

2020 ◽  
Vol 60 ◽  
pp. 96-102 ◽  
Author(s):  
Sidney Le ◽  
Emily Pellegrini ◽  
Abigail Green-Saxena ◽  
Charlotte Summers ◽  
Jana Hoffman ◽  
...  
Keyword(s):  
Machine Learning ◽  
Acute Respiratory Distress Syndrome ◽  
Respiratory Distress Syndrome ◽  
Respiratory Distress ◽  
Distress Syndrome ◽  
Supervised Machine Learning ◽  
Early Prediction

Alveolar Recruitment in Pulmonary and Extrapulmonary Acute Respiratory Distress Syndrome

2007 ◽  
Vol 106 (2) ◽  
pp. 212-217 ◽  
Author(s):  
Arnaud W. Thille ◽  
Jean-Christophe M. Richard ◽  
Salvatore M. Maggiore ◽  
V Marco Ranieri ◽  
Laurent Brochard
Keyword(s):  
Acute Respiratory Distress Syndrome ◽  
Respiratory Distress Syndrome ◽  
Respiratory Distress ◽  
Tidal Volume ◽  
Distress Syndrome ◽  
Alveolar Recruitment ◽  
High Peep ◽  
Static Compliance ◽  
Volume Curve ◽  
Pressure Volume Curve

Background Alveolar recruitment in response to positive end-expiratory pressure (PEEP) may differ between pulmonary and extrapulmonary acute respiratory distress syndrome (ARDS), and alveolar recruitment values may differ when measured by pressure-volume curve compared with static compliance. Methods The authors compared PEEP-induced alveolar recruitment in 71 consecutive patients identified in a database. Patients were classified as having pulmonary, extrapulmonary, or mixed/uncertain ARDS. Pressure-volume curves with and without PEEP were available for all patients, and pressure-volume curves with two PEEP levels were available for 44 patients. Static compliance was calculated as tidal volume divided by pressure change for tidal volumes of 400 and 700 ml. Recruited volume was measured at an elastic pressure of 15 cm H2O. Results Volume recruited by PEEP (10 +/- 3 cm H2O) was 223 +/- 111 ml in the pulmonary ARDS group (29 patients), 206 +/- 164 ml in the extrapulmonary group (16 patients), and 242 +/- 176 ml in the mixed/uncertain group (26 patients) (P = 0.75). At high PEEP (14 +/- 2 cmH2O, 44 patients), recruited volumes were also similar (P = 0.60). With static compliance, recruitment was markedly underestimated and was dependent on tidal volume (226 +/- 148 ml using pressure-volume curve, 95 +/- 185 ml for a tidal volume of 400 ml, and 23 +/- 169 ml for 700 ml; P < 0.001). Conclusion In a large sample of patients, classification of ARDS was uncertain in more than one third of patients, and alveolar recruitment was similar in pulmonary and extrapulmonary ARDS. PEEP levels should not be determined based on cause of ARDS.

Sign in / Sign up

Export Citation Format

Share Document