AI-based multi-modal integration (ScanCov scores) of clinical characteristics, lab tests and chest CTs improves COVID-19 outcome prediction of hospitalized patients
| Autor: | Simon Jégou, Jean-Philippe Lamarque, Matthieu Devilder, Paul Jehanno, Emilie Chouzenoux, Etienne Bendjebbar, Frank Chemouni, Sagar Verma, Nicolas Loiseau, Thomas Clozel, Franck Griscelli, Yannick Boursin, Imad Bousaid, Marie-Pauline Talabard, Elodie Pronier, Matthieu Terris, Tasnim Dardouri, Emmanuel Planchet, Olivier Dehaene, Jocelyn Dachary, Kathryn Schutte, Gilles Wainrib, Olivier Meyrignac, Corinne Balleyguier, Michael G. B. Blum, Yingping Li, Hugues Talbot, Mikael Azoulay, Fabien Brulport, Meriem Sefta, Fabrice Barlesi, Mansouria Merad, Remy Dubois, Samer Soliman, Paul Trichelair, Marie-France Bellin, Gabriel Garcia, Adrian Gonzalez, Jean-Christophe Pesquet, Nicolas Tetelboum, Samy Ammari, Nathalie Lassau, Ana Neacsu, Paul Herent, Jean-Baptiste Schiratti, Annabelle Stoclin, Hugo Gortais, Kavya Gupta |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
medicine.medical_specialty
Clinical variables medicine.diagnostic_test Coronavirus disease 2019 (COVID-19) business.industry Hospitalized patients Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Computed tomography 3. Good health 030218 nuclear medicine & medical imaging 03 medical and health sciences 0302 clinical medicine 030220 oncology & carcinogenesis Intensive care medicine Medical physics business Outcome prediction |
| DOI: | 10.1101/2020.05.14.20101972 |
| Popis: | With 15% of severe cases among hospitalized patients, the SARS-COV-2 pandemic has put tremendous pressure on Intensive Care Units, and made the identification of early predictors of future 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 future 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 more accurate predictions than using clinical and biological data alone. These findings show that CT scans provide insightful early predictors of future severity. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|