AI-based multi-modal integration of clinical characteristics, lab tests and chest CTs improves COVID-19 outcome prediction of hospitalized patients

Nathalie Lassau, et al.

Like Comment

Received Date: 19th May 20

Nathalie LassauSamy AmmariEmilie ChouzenouxHugo GortaisPaul HerentMatthieu DevilderSamer SolimanOlivier MeyrignacMarie-Pauline TalabardJean-Philippe LamarqueRemy DuboisNicolas LoiseauPaul TrichelairEtienne BendjebbarGabriel GarciaCorinne BalleyguierMansouria MeradAnnabelle StoclinSimon JegouFranck GriscelliNicolas TetelboumYingping LiSagar VermaMatthieu TerrisTasnim DardouriKavya GuptaAna NeacsuFrank ChemouniMeriem SeftaPaul JehannoImad BousaidYannick BoursinEmmanuel PlanchetMikael AzoulayJocelyn DacharyFabien BrulportAdrian GonzalezOlivier DehaeneJean-Baptiste SchirattiKathryn SchutteJean-Christophe PesquetHugues TalbotElodie PronierGilles Wainrib Thomas ClozelFabrice BarlesiMarie-France BellinMichael G. B. Blum

With 15% of severe cases among hospitalized patients1, the SARS-COV-2 pandemic has put tremendous pressure on Intensive Care Units, and made the identification of early predictors of severity a public health priority. We collected clinical and biological data, as well as CT scan images and radiology reports  from 1,003 coronavirus-infected patients from two French hospitals. Radiologists' manual CT annotations were also available. We first identified 11 clinical variables and 3 types of radiologist-reported features significantly associated with prognosis. Next, focusing on the CT images, we  trained deep learning models to automatically segment the scans and reproduce radiologists' annotations. We also built CT image-based deep learning models that predicted severity better than models based on the radiologists' scan reports. Finally, we showed that including CT scan features alongside the clinical and biological data yielded even more accurate predictions. These findings show that CT scans provide insightful early predictors of severity.

Read in full at medRxiv

This is an abstract of a preprint hosted on an independent third party site. It has not been peer reviewed but is currently under consideration at Nature Communications.

Nature Communications

Nature Research, Springer Nature