TY - JOUR
T1 - Assessing Trustworthy AI in Times of COVID-19.
T2 - Deep Learning for Predicting a Multiregional Score Conveying the Degree of Lung Compromise in COVID-19 Patients
AU - Allahabadi, Himanshi
AU - Amann, Julia
AU - Balot, Isabelle
AU - Beretta, Andrea
AU - Binkley, Charles
AU - Bozenhard, Jonas
AU - Bruneault, Frederick
AU - Brusseau, James
AU - Candemir, Sema
AU - Cappellini, Luca Alessandro
AU - Chakraborty, Subrata
AU - Cherciu, Nicoleta
AU - Cociancig, Christina
AU - Coffee, Megan
AU - Ek, Irene
AU - Espinosa-Leal, Leonardo
AU - Farina, Davide
AU - Fieux-Castagnet, Genevieve
AU - Frauenfelder, Thomas
AU - Gallucci, Alessio
AU - Giuliani, Guya
AU - Golda, Adam
AU - van Halem, Irmhild
AU - Hildt, Elisabeth
AU - Holm, Sune
AU - Kararigas, Georgios
AU - Krier, Sebastien A
AU - Kuhne, Ulrich
AU - Lizzi, Francesca
AU - Madai, Vince I
AU - Markus, Aniek F
AU - Masis, Serg
AU - Mathez, Emilie Wiinblad
AU - Mureddu, Francesco
AU - Neri, Emanuele
AU - Osika, Walter
AU - Ozols, Matiss
AU - Panigutti, Cecilia
AU - Parent, Brendan
AU - Pratesi, Francesca
AU - Moreno-Sanchez, Pedro A
AU - Sartor, Giovanni
AU - Savardi, Mattia
AU - Signoroni, Alberto
AU - Sormunen, Hanna-Maria
AU - Spezzatti, Andy
AU - Srivastava, Adarsh
AU - Stephansen, Annette F
AU - Theng, Lau Bee
AU - Tithi, Jesmin Jahan
AU - Tuominen, Jarno
AU - Umbrello, Steven
AU - Vaccher, Filippo
AU - Vetter, Dennis
AU - Westerlund, Magnus
AU - Wurth, Renee
AU - Zicari, Roberto V
N1 - ACKNOWLEDGMENT
The authors would like to thank Elia Belussi, Matthias Braun, Helga Brøgger, Andrew Bushell, Marcelo Corrales Compagnucci, Boris Düdder, Mads Kjolby, Maria Forss,
Fosca Giannotti, Thomas Gilbert, David Higgins, Asiatu Agnes Jalloh, Ahmed Khali, Federica Lucivero, Oriana Medlicott, Timo Minssen, Belona Sonna, Leonoor
Tideman, and Karsten Tolle who actively participated in Zoom meetings and/or offered valuable comments. Brendan Parent is the principal investigator on a Robert Wood Johnson
Foundation grant to study the ethics of big data for AI applications in health care.
PY - 2022/12
Y1 - 2022/12
N2 - 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.
AB - 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.
U2 - 10.1109/TTS.2022.3195114
DO - 10.1109/TTS.2022.3195114
M3 - Article
C2 - 36573115
VL - 3
SP - 272
EP - 289
IS - 4
ER -