A novel model to predict severe COVID-19 and mortality using an artificial intelligence algorithm to interpret chest radiographs and clinical variables

Abstract

BACKGROUNDPatients with COVID-19 could develop severe disease requiring admission to the Intensive Care Unit (ICU). This manuscript presents a novel method that predicts whether a patient will need admission to the ICU and assess the risk of in-hospital mortality by training a deep learning model that combines a set of clinical variables and features in the Chest-X-Rays.METHODSThis was a prospective diagnostic test study. Patients with confirmed SARS-CoV-2 infection between March 2020 and January 2021 were included. This study was designed to build predictive models obtained by training convolutional neural networks for Chest-X-ray images using an artificial intelligence (AI) tool and a Random Forest analysis to identify critical clinical variables. Then, both architectures were connected and fine-tuned to provide combined models.RESULTSA total of 2552 patients were included in the clinical cohort. The variables independently associated with ICU admission were age, the fraction of inspired oxygen - FiO2 on admission, dyspnoea on admission, and obesity. Moreover, the variables associated with hospital mortality were age, the fraction of inspired oxygen - FiO2 on admission, and dyspnoea. When implementing the AI model to interpret the Chest-X-rays and the clinical variable identified by random forest, we developed a model that accurately predicts ICU admission (AUC:0.92±0.04) and hospital mortality (AUC:0.81±0.06) in patients with confirmed COVID-19.CONCLUSIONSThis automated Chest-X-ray interpretation algorithm, along with clinical variables, is a reliable alternative to identify patients at risk of developing severe COVID-19 that might require admission to the ICU.

Rapid Assessment of COVID-19 Mortality Risk with GASS Classifiers

“…Such findings are consistent with previous studies: male sex has been already linked to poorer chances of survival from COVID-19 and with increased risk of admission to the Intensive Care Units (ICUs) [ 30 ]; similarly, the PaO2/FiO2 ratio was already shown to be independently associated with worse COVID-19 outcomes [ 29 , 31 ]. Furthermore, recent studies showed how COVID-19 inpatients present commonly with laboratory abnormalities and the role of inflammatory, and organ-damage markers has been also consistently reported [ 32 , 33 ].…”

“…Importantly, the SVM22-GASS bases its prediction on the analysis of clinical variables that can be quickly and inexpensively retrieved early on during hospital admission. This is a significant advantage over other popular multivariate methods that rely on the manual analysis of medical imaging results (e.g., [ 29 ]), and yet achieve comparable predictive performance (validation AUC = 0.88). Like most machine-learning-based classifiers, the SVM22-GASS, the 4C and the Piacenza scores are purely data-driven.…”

“…Predicting the mortality risk of hospitalized COVID-19 patients could go a long way to optimal allocation of hospital resources. Popular clinical scores are potentially useful in this regard, but have proven to be less accurate than COVID-19-specific methods [ 10 , 29 ]. In this work we have introduced and validated two new methods for predicting 30-day mortality risk: the Clinical-GASS and the SVM22-GASS.…”

“…Popular clinical scores of general applicability, such as the Charlson Comorbidity Score (CCI) [ 4 ], the modified Elixhauser Index (mEI) [ 43 ], the National Early Warning Score 2 (NEWS2) [ 44 ] and the CURB models [ 8 , 9 ] have been used during the current pandemic to predict COVID-19 outcomes with discrete success. However, they tend to underperform COVID-19-specific clinical scores [ 10 , 29 ].…”

Subgroup State Prediction under Different Noise Levels Using MODWT and XGBoost

“…Wang et al [ 25 ] introduced a weight decay random forest model to achieve ICU readmission classification based on sparse data and integrated the missing value analysis and the likelihood ratio test for the distribution characteristics of time series indicators. Munera et al [ 26 ] applied a random forest model to predict the probability of ICU admission or hospital mortality, which was combined with a logistic regression model designed to select the clinical variables and laboratory results that best predicted the outcomes. Varghese et al [ 27 ] employed the AdaBoost classifier to achieve early identification of COVID-19 patients at risk of a poor prognosis as defined by the need for ICU and mechanical ventilation, and showed the favourable performance of the AdaBoost classifier in the prediction.…”