Full metadata record
DC FieldValueLanguage
dc.creatorBermejo-Peláez, D. (David)-
dc.creatorSan-José-Estépar, R. (Raúl)-
dc.creatorFernández-Velilla, M. (María)-
dc.creatorPalacios-Miras, C. (Carmelo)-
dc.creatorGallardo-Madueño, G. (Guillermo)-
dc.creatorBenegas, M. (Mariana)-
dc.creatorGotera-Rivera, C. (Carolina)-
dc.creatorCuerpo, S. (Sara)-
dc.creatorLuengo-Oroz, M. (Miguel)-
dc.creatorSellarés, J. (Jacobo)-
dc.creatorSánchez, M. (Marcelo)-
dc.creatorBastarrika, G. (Gorka)-
dc.creatorPeces-Barba, G. (German)-
dc.creatorSeijo, L. (Luis)-
dc.creatorLedesma-Carbayo, M.J. (María J.)-
dc.identifier.citationBermejo-Peláez, D.; San José Estépar, R.; Fernández-Velilla, M.; et al. "Deep learning-based lesion subtyping and prediction of clinical outcomes in COVID-19 pneumonia using chest CT". Scientific Reports. 12 (1), 2022, 9387es
dc.description.abstractThe main objective of this work is to develop and evaluate an artificial intelligence system based on deep learning capable of automatically identifying, quantifying, and characterizing COVID-19 pneumonia patterns in order to assess disease severity and predict clinical outcomes, and to compare the prediction performance with respect to human reader severity assessment and whole lung radiomics. We propose a deep learning based scheme to automatically segment the different lesion subtypes in nonenhanced CT scans. The automatic lesion quantification was used to predict clinical outcomes. The proposed technique has been independently tested in a multicentric cohort of 103 patients, retrospectively collected between March and July of 2020. Segmentation of lesion subtypes was evaluated using both overlapping (Dice) and distance-based (Hausdorff and average surface) metrics, while the proposed system to predict clinically relevant outcomes was assessed using the area under the curve (AUC). Additionally, other metrics including sensitivity, specificity, positive predictive value and negative predictive value were estimated. 95% confidence intervals were properly calculated. The agreement between the automatic estimate of parenchymal damage (%) and the radiologists' severity scoring was strong, with a Spearman correlation coefficient (R) of 0.83. The automatic quantification of lesion subtypes was able to predict patient mortality, admission to the Intensive Care Units (ICU) and need for mechanical ventilation with an AUC of 0.87, 0.73 and 0.68 respectively. The proposed artificial intelligence system enabled a better prediction of those clinically relevant outcomes when compared to the radiologists' interpretation and to whole lung radiomics. In conclusion, deep learning lesion subtyping in COVID-19 pneumonia from noncontrast chest CT enables quantitative assessment of disease severity and better prediction of clinical outcomes with respect to whole lung radiomics or radiologists' severity score.-
dc.description.sponsorshipThis work was partially funded by projects RTI2018-098682-B-I00 (MCIU/AEI/FEDER,UE) by the Spanish ministry of Science and Innovation, and the initiative REACT-EU Comunidad de Madrid both cofunded by the European Union ERDF (European Regional Development Fund), as well as the Leonardo grant to researchers and cultural creators 2019 from Fundación BBVA. Raúl San José Estépar was supported by NIH grant R21LM013670. Research work of Dr. Jacobo Sellarés is financed by the grant SLT008/18/00176 and the support of the Department of Health of the Generalitat de Catalunya, in the call for grants 2019–2021, as part of the Strategic Research and Innovation Plan in Health 2016–2020. It has also been financed by FEDER Funds (PI19/01152), SEPAR (PI17/562 & PI18/792), SOCAP, FUCAP and Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS).-
dc.subjectÁrea de Medicina Clínica y Epidemiología-
dc.titleDeep learning-based lesion subtyping and prediction of clinical outcomes in COVID-19 pneumonia using chest CT-
dc.description.noteThis article is licensed under a Creative Commons Attribution 4.0 International License-
dadun.citation.publicationNameScientific Reports-

Files in This Item:
1.67 MB
Adobe PDF

Statistics and impact
0 citas en
0 citas en

Items in Dadun are protected by copyright, with all rights reserved, unless otherwise indicated.