Integrating deep learning CT-scan model, biological and clinical variables to predict severity of COVID-19 patients
Nathalie Lassau, Samy Ammari, Émilie Chouzenoux, Hugo Gortais, Paul Hérent, Matthieu Devilder, Samer Soliman, Olivier Meyrignac, Marie-Pauline Talabard, Jean-Philippe Lamarque, Rémy Dubois, Nicolas Loiseau, Paul Trichelair, Etienne Bendjebbar, Gabriel Garcia, Corinne Balleyguier, Mansouria Merad, Annabelle Stoclin, Simon Jégou, Frank Griscelli, Nicolas Tetelboum, Yingping Li, Sagar Verma, Matthieu Terris, Tasnim Dardouri, Kavya Gupta, Ana Neacşu, Frank Chemouni, Meriem Sefta, Paul Jehanno, Imad Bousaid, Yannick Boursin, Emmanuel Planchet, Mikael Azoulay, Jocelyn Dachary, Fabien Brulport, Adrian Gonzalez, Olivier Dehaene, Jean-Baptiste Schiratti, Kathryn Schutte, Jean‐Christophe Pesquet, Hugues Talbot, Elodie Pronier, Gilles Wainrib, Thomas Clozel, Fabrice Barlési, Marie‐France Bellin, Michaël G. B. Blum
Abstract
The SARS-COV-2 pandemic has put pressure on intensive care units, so that identifying predictors of disease severity is a priority. We collect 58 clinical and biological variables, and chest CT scan data, from 1003 coronavirus-infected patients from two French hospitals. We train a deep learning model based on CT scans to predict severity. We then construct the multimodal AI-severity score that includes 5 clinical and biological variables (age, sex, oxygenation, urea, platelet) in addition to the deep learning model. We show that neural network analysis of CT-scans brings unique prognosis information, although it is correlated with other markers of severity (oxygenation, LDH, and CRP) explaining the measurable but limited 0.03 increase of AUC obtained when adding CT-scan information to clinical variables. Here, we show that when comparing AI-severity with 11 existing severity scores, we find significantly improved prognosis performance; AI-severity can therefore rapidly become a reference scoring approach.