Exécution de notre outil de synthèse sur un nouvel article

Vérifiez l’article ci-dessous, puis lancez sa synthèse.


Assessing Trustworthy AI in Times of COVID-19: Deep Learning for Predicting a Multiregional Score Conveying the Degree of Lung Compromise in COVID-19 Patients

Himanshi Allahabadi, Julia Amann (ETH Zurich), Isabelle Balot, Andrea Beretta (Istituto di Scienza e Tecnologie dell'Informazione "Alessandro Faedo", National Research Council), Charles E. Binkley (Hackensack Meridian Health), Jonas Bozenhard (University of Oxford), Frédérick Bruneault (Cégep André Laurendeau), James Brusseau (Pace University), Sema Candemir (The Ohio State University Wexner Medical Center, The Ohio State University), Luca Alessandro Cappellini (Humanitas University, IRCCS Humanitas Research Hospital), Subrata Chakraborty (University of Technology Sydney, University of New England), Nicoleta Cherciu (Scuola Superiore Sant'Anna), Christina Cociancig (University of Bremen), Megan P. Coffee (New York University), Irene Ek, Leonardo Espinosa-Leal (Arcada University of Applied Sciences), Davide Farina (University of Brescia), Geneviève Fieux-Castagnet (Société Nationale des Chemins de Fer Français (France)), Thomas Frauenfelder (University Hospital Zurich), Alessio Gallucci (Eindhoven University of Technology), Guya Giuliani (Ericsson (Sweden)), Adam Gołda, Irmhild van Halem, Elisabeth Hildt (Illinois Institute of Technology), Sune Holm (University of Copenhagen), Georgios Kararigas (University of Iceland), Sébastien Krier (Stanford University), Ulrich Kuhne, Francesca Lizzi (Scuola Normale Superiore), Vince I. Madai (Birmingham City University, Berlin Institute of Health at Charité - Universitätsmedizin Berlin), Aniek F. Markus (Erasmus MC), Serg Masis (Syngenta (United States)), Emilie Wiinblad Mathez, Francesco Mureddu, Emanuele Neri (University of Pisa), Walter Osika (Karolinska Institutet), Matiss Ozols (University of Cambridge, Wellcome Sanger Institute), Cecilia Panigutti (University of Pisa), Brendan Parent (New York University), Francesca Pratesi (Istituto di Scienza e Tecnologie dell'Informazione "Alessandro Faedo", National Research Council), Pedro A. Moreno-Sánchez (Seinäjoki University of Applied Sciences), Giovanni Sartor B (European University Institute, University of Bologna), Mattia Savardi (University of Brescia), Alberto Signoroni (University of Brescia), Hanna-Maria Sormunen, Andy Spezzatti (University of California, Berkeley), Adarsh Srivastava, Annette F. Stephansen (NORCE Research AS), Lau Bee Theng (Swinburne University of Technology Sarawak Campus), Jesmin Jahan Tithi (Intel (United States), Stony Brook University), Jarno Tuominen (University of Turku), Steven Umbrello (Delft University of Technology), Filippo Vaccher (University of Brescia), Dennis Vetter (Goethe University Frankfurt), Magnus Westerlund (Arcada University of Applied Sciences, Høyskolen Kristiania), Renee C. Wurth (Harvard University), Roberto V. Zicari (Seoul National University, Arcada University of Applied Sciences)

IEEE Transactions on Technology and Society
Licence : non indiquée (tous droits réservés)

Résumé : This article's main contributions are twofold: 1) to demonstrate how to apply the general European Union's High-Level Expert Group's (EU HLEG) guidelines for trustworthy AI in practice for the domain of healthcare and 2) to investigate the research question of what does "trustworthy AI" mean at the time of the COVID-19 pandemic. To this end, we present the results of a post-hoc self-assessment to evaluate the trustworthiness of an AI system for predicting a multiregional score conveying the degree of lung compromise in COVID-19 patients, developed and verified by an interdisciplinary team with members from academia, public hospitals, and industry in time of pandemic. The AI system aims to help radiologists to estimate and communicate the severity of damage in a patient's lung from Chest X-rays. It has been experimentally deployed in the radiology department of the ASST Spedali Civili clinic in Brescia, Italy, since December 2020 during pandemic time. The methodology we have applied for our post-hoc assessment, called Z-Inspection®, uses sociotechnical scenarios to identify ethical, technical, and domain-specific issues in the use of the AI system in the context of the pandemic.

Publié le 29 Jul. 2022