Deep learning of chest X‑rays can predict mechanical ventilation outcome in ICU‑admitted COVID‑19 patients

Authors: Gourdeau, DanielPotvin, OlivierBiem, Jason HenryCloutier, Florence; Abrougui, Lyna; Archambault, Patrick; Chartrand‑Lefebvre, Carl; Dieumegarde, LouisGagné, ChristianGagnon, Louis; Giguère, Raphaelle; Hains, AlexandreLe-Khac, Huy.Lemieux, SimonLévesque, Marie-Hélène; Nepveu, Simon; Rosenbloom, Lorne; Tang, An; Yang, Issac; Duchesne, NathalieDuchesne, Simon
Abstract: The COVID-19 pandemic repeatedly overwhelms healthcare systems capacity and forced the development and implementation of triage guidelines in ICU for scarce resources (e.g. mechanical ventilation). These guidelines were often based on known risk factors for COVID-19. It is proposed that image data, specifcally bedside computed X-ray (CXR), provide additional predictive information on mortality following mechanical ventilation that can be incorporated in the guidelines. Deep transfer learning was used to extract convolutional features from a systematically collected, multiinstitutional dataset of COVID-19 ICU patients. A model predicting outcome of mechanical ventilation (remission or mortality) was trained on the extracted features and compared to a model based on known, aggregated risk factors. The model reached a 0.702 area under the curve (95% CI 0.707-0.694) at predicting mechanical ventilation outcome from pre-intubation CXRs, higher than the risk factor model. Combining imaging data and risk factors increased model performance to 0.743 AUC (95% CI 0.746-0.732). Additionally, a post-hoc analysis showed an increase performance on high-quality than low-quality CXRs, suggesting that using only high-quality images would result in an even stronger model.
Document Type: Article de recherche
Issue Date: 13 April 2022
Open Access Date: 10 May 2022
Document version: VoR
Creative Commons Licence:
This document was published in: Scientific Reports, Vol. 12 (2022)
Springer Nature
Alternative version: 10.1038/s41598-022-10136-9
Collection:Articles publiés dans des revues avec comité de lecture

Files in this item:
Description SizeFormat 
Gourdeau_et_al-2022B-Scientific_Reports.pdf1.6 MBAdobe PDFThumbnail
All documents in CorpusUL are protected by Copyright Act of Canada.