eARDS: A multi-center validation of an interpretable machine learning algorithm of early onset Acute Respiratory Distress Syndrome (ARDS) among critically ill adults with COVID-19

Abstract

We present an interpretable machine learning algorithm called ‘eARDS’ for predicting ARDS in an ICU population comprising COVID-19 patients, up to 12-hours before satisfying the Berlin clinical criteria. The analysis was conducted on data collected from the Intensive care units (ICU) at Emory Healthcare, Atlanta, GA and University of Tennessee Health Science Center, Memphis, TN and the Cerner® Health Facts Deidentified Database, a multi-site COVID-19 EMR database. The participants in the analysis consisted of adults over 18 years of age. Clinical data from 35,804 patients who developed ARDS and controls were used to generate predictive models that identify risk for ARDS onset up to 12-hours before satisfying the Berlin criteria. We identified salient features from the electronic medical record that predicted respiratory failure among this population. The machine learning algorithm which provided the best performance exhibited AUROC of 0.89 (95% CI = 0.88–0.90), sensitivity of 0.77 (95% CI = 0.75–0.78), specificity 0.85 (95% CI = 085–0.86). Validation performance across two separate health systems (comprising 899 COVID-19 patients) exhibited AUROC of 0.82 (0.81–0.83) and 0.89 (0.87, 0.90). Important features for prediction of ARDS included minimum oxygen saturation (SpO2), standard deviation of the systolic blood pressure (SBP), O2 flow, and maximum respiratory rate over an observational window of 16-hours. Analyzing the performance of the model across various cohorts indicates that the model performed best among a younger age group (18–40) (AUROC = 0.93 [0.92–0.94]), compared to an older age group (80+) (AUROC = 0.81 [0.81–0.82]). The model performance was comparable on both male and female groups, but performed significantly better on the severe ARDS group compared to the mild and moderate groups. The eARDS system demonstrated robust performance for predicting COVID19 patients who developed ARDS at least 12-hours before the Berlin clinical criteria, across two independent health systems.

Clinical prediction system of complications among patients with COVID-19: A development and validation retrospective multicentre study during first wave of the pandemic

“…Although the results are not directly comparable due to differences in study design, our system achieved 0.91 and 0.89 AUROC in test sets A & B, respectively, without needing any imaging or laboratory-test results. In another study, an ARDS prediction model achieved 0.89 AUROC using patient demographics, interventions, comorbidities, 17 laboratory-test results and eight vital signs [ 33 ]. In comparison, our system achieved 0.85 AUROC in test set A and 0.83 AUROC in test set B.…”

Management of COVID-19-Associated Acute Respiratory Failure with Alternatives to Invasive Mechanical Ventilation: High-Flow Oxygen, Continuous Positive Airway Pressure, and Noninvasive Ventilation

“…Hence, and perhaps not surprising, several different respiratory measures seem to contribute substantially to predicting ARDS in patients with COVID-19. However, as a possible future biomarker, the lowest measured platelet count also contributed to the algorithm [29] In patients without COVID-19, several biomarkers seem promising in diagnosing and predicting the development of ARF, and some have already to some degree proven their value. Plasma surfactant protein D (SP-D) has been shown to increase within 48 h of admission to ICU in patients who developed ARDS and to predict the long-term need for IMV and mortality [30][31][32], and angiopoietin-2 was able to predict pulmonary affection in cohort studies in critically ill patients with various underlying courses [33][34][35][36] and also predict severity of illness and mortality [37][38][39][40].…”

“…Hence, and perhaps not surprising, several different respiratory measures seem to contribute substantially to predicting ARDS in patients with COVID-19. However, as a possible future biomarker, the lowest measured platelet count also contributed to the algorithm [ 29 ]…”

Machine Learning Approaches to Identify Discriminative Signatures of Volatile Organic Compounds (VOCs) from Bacteria and Fungi Using SPME-DART-MS

“…In recent years, applications of machine learning (ML) and artificial intelligence have shown great promise in advancing the field of healthcare and critical care [ 23 ]. ML models are able to identify physiomarkers that help in early detection of sepsis [ 24 ] and predict life-threatening conditions such as acute respiratory distress syndrome (ARDS) using ICU data [ 25 ] and gene expression signatures [ 26 ]. A major area of research currently is early and accurate detection of infections from microbial VOCs and several statistical and machine learning methods have been successfully developed to this end [ 4 , 27 ].…”