Mohammed Sayed1, David Riaño2, Jesús Villar3,4,5. 1. Banzai Research Group On Artificial Intelligence, Department of Computer Engineering, Universitat Rovira I Virgili, Av Paisos Catalans 26, 43007, Tarragona, Spain. mgamal.sayed@urv.cat. 2. Banzai Research Group On Artificial Intelligence, Department of Computer Engineering, Universitat Rovira I Virgili, Av Paisos Catalans 26, 43007, Tarragona, Spain. david.riano@urv.cat. 3. Centro de Investigación Biomédica en Red de Enfermedades Respiratorias, Instituto de Salud Carlos III, Madrid, Spain. jesus.villar54@gmail.com. 4. Multidisciplinary Organ Dysfunction Evaluation Research Network, Research Unit, Hospital Universitario Dr Negrín, Barranco de la Ballena s/n, 4th Floor -South Wing, 35019, Las Palmas de Gran Canaria, Spain. jesus.villar54@gmail.com. 5. Keenan Research Center for Biomedical Science at the Li Ka Shing Knowledge Institute, St Michael's Hospital, Toronto, ON, Canada. jesus.villar54@gmail.com.
Abstract
BACKGROUND: Usually, arterial oxygenation in patients with the acute respiratory distress syndrome (ARDS) improves substantially by increasing the level of positive end-expiratory pressure (PEEP). Herein, we are proposing a novel variable [PaO2/(FiO2xPEEP) or P/FPE] for PEEP ≥ 5 to address Berlin's definition gap for ARDS severity by using machine learning (ML) approaches. METHODS: We examined P/FPE values delimiting the boundaries of mild, moderate, and severe ARDS. We applied ML to predict ARDS severity after onset over time by comparing current Berlin PaO2/FiO2 criteria with P/FPE under three different scenarios. We extracted clinical data from the first 3 ICU days after ARDS onset (N = 2738, 1519, and 1341 patients, respectively) from MIMIC-III database according to Berlin criteria for severity. Then, we used the multicenter database eICU (2014-2015) and extracted data from the first 3 ICU days after ARDS onset (N = 5153, 2981, and 2326 patients, respectively). Disease progression in each database was tracked along those 3 ICU days to assess ARDS severity. Three robust ML classification techniques were implemented using Python 3.7 (LightGBM, RF, and XGBoost) for predicting ARDS severity over time. RESULTS: P/FPE ratio outperformed PaO2/FiO2 ratio in all ML models for predicting ARDS severity after onset over time (MIMIC-III: AUC 0.711-0.788 and CORR 0.376-0.566; eICU: AUC 0.734-0.873 and CORR 0.511-0.745). CONCLUSIONS: The novel P/FPE ratio to assess ARDS severity after onset over time is markedly better than current PaO2/FiO2 criteria. The use of P/FPE could help to manage ARDS patients with a more precise therapeutic regimen for each ARDS category of severity.
BACKGROUND: Usually, arterial oxygenation in patients with the acute respiratory distress syndrome (ARDS) improves substantially by increasing the level of positive end-expiratory pressure (PEEP). Herein, we are proposing a novel variable [PaO2/(FiO2xPEEP) or P/FPE] for PEEP ≥ 5 to address Berlin's definition gap for ARDS severity by using machine learning (ML) approaches. METHODS: We examined P/FPE values delimiting the boundaries of mild, moderate, and severe ARDS. We applied ML to predict ARDS severity after onset over time by comparing current Berlin PaO2/FiO2 criteria with P/FPE under three different scenarios. We extracted clinical data from the first 3 ICU days after ARDS onset (N = 2738, 1519, and 1341 patients, respectively) from MIMIC-III database according to Berlin criteria for severity. Then, we used the multicenter database eICU (2014-2015) and extracted data from the first 3 ICU days after ARDS onset (N = 5153, 2981, and 2326 patients, respectively). Disease progression in each database was tracked along those 3 ICU days to assess ARDS severity. Three robust ML classification techniques were implemented using Python 3.7 (LightGBM, RF, and XGBoost) for predicting ARDS severity over time. RESULTS:P/FPE ratio outperformed PaO2/FiO2 ratio in all ML models for predicting ARDS severity after onset over time (MIMIC-III: AUC 0.711-0.788 and CORR 0.376-0.566; eICU: AUC 0.734-0.873 and CORR 0.511-0.745). CONCLUSIONS: The novel P/FPE ratio to assess ARDS severity after onset over time is markedly better than current PaO2/FiO2 criteria. The use of P/FPE could help to manage ARDSpatients with a more precise therapeutic regimen for each ARDS category of severity.
Authors: Carolyn S Calfee; Kevin Delucchi; Polly E Parsons; B Taylor Thompson; Lorraine B Ware; Michael A Matthay Journal: Lancet Respir Med Date: 2014-05-19 Impact factor: 30.700
Authors: Giacomo Bellani; John G Laffey; Tài Pham; Eddy Fan; Laurent Brochard; Andres Esteban; Luciano Gattinoni; Frank van Haren; Anders Larsson; Daniel F McAuley; Marco Ranieri; Gordon Rubenfeld; B Taylor Thompson; Hermann Wrigge; Arthur S Slutsky; Antonio Pesenti Journal: JAMA Date: 2016-02-23 Impact factor: 56.272
Authors: G R Bernard; A Artigas; K L Brigham; J Carlet; K Falke; L Hudson; M Lamy; J R LeGall; A Morris; R Spragg Journal: Intensive Care Med Date: 1994 Impact factor: 17.440
Authors: Niall D Ferguson; Eddy Fan; Luigi Camporota; Massimo Antonelli; Antonio Anzueto; Richard Beale; Laurent Brochard; Roy Brower; Andrés Esteban; Luciano Gattinoni; Andrew Rhodes; Arthur S Slutsky; Jean-Louis Vincent; Gordon D Rubenfeld; B Taylor Thompson; V Marco Ranieri Journal: Intensive Care Med Date: 2012-08-25 Impact factor: 17.440
Authors: Pratik Sinha; Kevin L Delucchi; B Taylor Thompson; Daniel F McAuley; Michael A Matthay; Carolyn S Calfee Journal: Intensive Care Med Date: 2018-10-05 Impact factor: 17.440